INDEX

Explanations

negations or contradictions

oai_token-act-pair · gpt-3.5-turbo

negations or phrases that downplay threats or concerns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 defaults

-0.73

Ĥİ

-0.70

©¶æ

-0.66

 assurances

-0.62

 impunity

-0.62

Ń·

-0.60

¿½

-0.60

orde

-0.56

®

-0.55

 shortcuts

-0.54

POSITIVE LOGITS

existent

1.04

hin

0.98

 exist

0.97

 mattered

0.89

 existed

0.87

ogether

0.85

 matters

0.84

 even

0.83

 bother

0.82

 matter

0.79

Activations Density 0.282%

negations or contradictions

negations or phrases that downplay threats or concerns

No Comments

No Known Activations