INDEX

Explanations

mentions of the word "drama."

oai_token-act-pair · gpt-3.5-turbo

references to drama 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ovo

-0.79

ibel

-0.77

anza

-0.74

erers

-0.73

achev

-0.73

itude

-0.71

ees

-0.70

ossier

-0.69

olphin

-0.68

imates

-0.68

POSITIVE LOGITS

 queens

0.96

 drama

0.90

 dramas

0.87

 unfold

0.85

 queen

0.81

 unfolds

0.80

 involving

0.75

 comed

0.73

 Reign

0.73

 unfolded

0.72

Activations Density 0.020%

mentions of the word "drama."

references to drama in various contexts

No Comments

No Known Activations