INDEX

Explanations

mentions of individuals named Edward

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Edward" 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 @ 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

alore

-0.72

pora

-0.71

nda

-0.70

Äĩ

-0.69

awaru

-0.68

BOOK

-0.68

à¤

-0.65

lyak

-0.65

packing

-0.64

asta

-0.63

POSITIVE LOGITS

 Snowden

1.05

 Edward

0.87

 Cullen

0.86

 Byrne

0.84

sson

0.83

 Everett

0.79

 Neville

0.76

 Nero

0.76

Gib

0.74

 Norton

0.73

Activations Density 0.009%

mentions of individuals named Edward

mentions of the name "Edward" in various contexts

No Comments

No Known Activations