INDEX

Explanations

terms related to job-related information, qualifications, and categories of workers

oai_token-act-pair · gpt-3.5-turbo

terms associated with employment and organizational structures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cknow

-0.61

 skelet

-0.57

abus

-0.55

OLOG

-0.51

Spl

-0.51

raz

-0.51

atever

-0.51

dig

-0.51

icho

-0.51

dit

-0.51

POSITIVE LOGITS

 varies

1.29

 reflects

1.18

 differs

1.17

 involves

1.13

 constitutes

1.12

 depends

1.12

 tends

1.12

 coincides

1.12

 underscores

1.12

 illustrates

1.09

Activations Density 0.400%

terms related to job-related information, qualifications, and categories of workers

terms associated with employment and organizational structures

No Comments

No Known Activations

terms related to job-related information, qualifications, and categories of workers

terms associated with employment and organizational structures

No Comments

No Known Activations