INDEX

Explanations

the word "handful."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "handful."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

sein

-0.68

Train

-0.67

abolic

-0.66

perse

-0.63

iola

-0.63

ogens

-0.61

ACTED

-0.60

hemat

-0.60

poses

-0.59

ruption

-0.59

POSITIVE LOGITS

eenth

1.01

 dozen

0.93

een

0.90

uously

0.87

 thousand

0.85

dozen

0.83

 hundred

0.78

of

0.75

 thereof

0.74

 ILCS

0.74

Activations Density 0.017%

the word "handful."

occurrences of the word "handful."

No Comments

No Known Activations