INDEX

Explanations

numerical values or mentions of experimental results in scientific contexts

oai_token-act-pair · gpt-3.5-turbo

references to scientific classifications or terminologies

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

manship

-0.73

ãĤ¬

-0.67

ãĥ©ãĥ³

-0.64

 Bung

-0.63

 tongues

-0.62

geries

-0.60

 este

-0.60

destroy

-0.60

ilda

-0.59

ãĥŀ

-0.58

POSITIVE LOGITS

].

0.89

],[

0.85

–

0.80

â

0.80

Î²

0.80

âĢĲ

0.79

doi

0.77

 âĩ

0.77

)].

0.77

reviewed

0.76

Activations Density 0.138%

numerical values or mentions of experimental results in scientific contexts

references to scientific classifications or terminologies

No Comments

No Known Activations