INDEX

Explanations

actions or tasks that are perceived as challenging or difficult

oai_token-act-pair · gpt-3.5-turbo

verbs associated with difficulty and understanding tasks or concepts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 teasp

-0.60

 entitled

-0.59

bon

-0.58

 unspecified

-0.56

shall

-0.55

sis

-0.55

olon

-0.55

 Moran

-0.54

 proceeds

-0.53

OOOOOOOO

-0.53

POSITIVE LOGITS

 oneself

1.00

 nowadays

0.92

due

0.87

 accurately

0.79

 unless

0.78

 yourself

0.77

ensical

0.75

 visually

0.74

 compared

0.73

 reliably

0.72

Activations Density 0.187%

actions or tasks that are perceived as challenging or difficult

verbs associated with difficulty and understanding tasks or concepts

No Comments

No Known Activations