INDEX

Explanations

mentions of expertise or specialization

oai_token-act-pair · gpt-3.5-turbo

terms related to experience

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

 Crom

-0.69

 Cock

-0.69

pul

-0.68

 Hook

-0.68

 thumbs

-0.61

 Luck

-0.60

 Hipp

-0.59

 Marble

-0.59

 rough

-0.58

Dew

-0.58

POSITIVE LOGITS

ropri

1.54

orter

1.53

orters

1.52

anse

1.47

ository

1.33

atri

1.31

onents

1.26

iring

1.26

resso

1.22

iry

1.16

Activations Density 0.012%

mentions of expertise or specialization

terms related to experience

No Comments

No Known Activations