INDEX

Explanations

phrases that denote quantity or comparison

oai_token-act-pair · gpt-3.5-turbo

references to multiple instances or occurrences of similar items or events

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

 Akin

-0.80

matter

-0.72

Emb

-0.72

 john

-0.67

 Ezek

-0.66

 mitochond

-0.63

Lov

-0.63

embed

-0.62

 Amon

-0.60

borough

-0.60

POSITIVE LOGITS

 quir

0.78

 finalists

0.76

 additions

0.70

 favorites

0.69

oggle

0.69

 contenders

0.66

renches

0.66

quished

0.63

 factors

0.63

phabet

0.63

Activations Density 0.077%

phrases that denote quantity or comparison

references to multiple instances or occurrences of similar items or events

No Comments

No Known Activations