INDEX

Explanations

phrases related to occupational or task-related roles

oai_token-act-pair · gpt-3.5-turbo

references to roles and responsibilities in various contexts

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

ongo

-0.62

ento

-0.61

Interstitial

-0.59

iasm

-0.58

cluding

-0.58

iolet

-0.58

OUP

-0.58

taboola

-0.56

 collaborations

-0.56

eele

-0.56

POSITIVE LOGITS

of

1.03

 vacated

0.95

 himself

0.93

iest

0.86

 willingly

0.85

 entrusted

0.83

 reluctantly

0.81

 anew

0.81

wrong

0.79

 convinc

0.78

Activations Density 0.240%

phrases related to occupational or task-related roles

references to roles and responsibilities in various contexts

No Comments

No Known Activations