INDEX

Explanations

mentions of achievements, roles, or significant events related to individuals

oai_token-act-pair · gpt-3.5-turbo

terms related to personal circumstances and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 oneself

-0.56

orks

-0.55

 Sark

-0.52

gr

-0.52

Ã°

-0.52

 Guan

-0.51

È

-0.51

 McDonnell

-0.51

uther

-0.51

Dow

-0.50

POSITIVE LOGITS

 counterparts

0.96

 counterpart

0.79

 brethren

0.78

 holdings

0.78

 selves

0.74

 predecessors

0.70

 portfolio

0.70

 cousins

0.69

 foray

0.69

 buddies

0.69

Activations Density 1.349%

mentions of achievements, roles, or significant events related to individuals

terms related to personal circumstances and experiences

No Comments

No Known Activations