INDEX

Explanations

mentions of the name "Edward Snowden"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Edward Snowden."

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

footed

-0.73

practice

-0.69

roads

-0.67

cone

-0.64

tackle

-0.63

sites

-0.62

unity

-0.61

stakes

-0.60

chairs

-0.59

sect

-0.58

POSITIVE LOGITS

 Snowden

1.19

 Cullen

0.89

 Byrne

0.87

mund

0.85

 Nero

0.82

 Norton

0.81

 Said

0.81

 Abbey

0.80

 Everett

0.78

IBLE

0.77

Activations Density 0.016%

mentions of the name "Edward Snowden"

mentions of the name "Edward Snowden."

No Comments

No Known Activations