INDEX

Explanations

mentions of the word "Pixar"

oai_token-act-pair · gpt-3.5-turbo

references to Pixar

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lehem

-0.74

¥µ

-0.72

ufact

-0.71

TPS

-0.68

riter

-0.67

udder

-0.67

 behavi

-0.66

INST

-0.66

iage

-0.66

à¨

-0.65

POSITIVE LOGITS

es

1.10

avier

0.96

eus

0.95

xes

0.94

xus

0.91

iang

0.89

edo

0.89

uries

0.89

iao

0.88

xon

0.85

Activations Density 0.027%

mentions of the word "Pixar"

references to Pixar

No Comments

No Known Activations