INDEX

Explanations

phrases related to self-talk and internal dialogue

oai_token-act-pair · gpt-3.5-turbo

concepts related to self-reflection and personal actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Reconstruction

-0.64

ighed

-0.57

 purported

-0.56

gerald

-0.55

 supplemented

-0.55

 Annex

-0.54

 instituted

-0.51

Ĭ±

-0.50

Il

-0.50

-0.49

POSITIVE LOGITS

 yourself

1.50

 yourselves

1.27

 Yourself

1.09

 your

0.96

 YOUR

0.85

your

0.78

Your

0.76

 Your

0.72

poke

0.64

 yours

0.63

Activations Density 0.743%

phrases related to self-talk and internal dialogue

concepts related to self-reflection and personal actions

No Comments

No Known Activations