INDEX

Explanations

names of sports teams or clubs

oai_token-act-pair · gpt-3.5-turbo

names of football clubs and teams

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

schild

-0.76

Zup

-0.74

ãĥ´

-0.73

haar

-0.72

enegger

-0.71

clus

-0.70

 narration

-0.70

 Ples

-0.70

pees

-0.70

ographer

-0.69

POSITIVE LOGITS

 Wolves

0.99

 Hots

0.98

 Athletic

0.96

FC

0.96

 midfielder

0.92

 Sporting

0.89

wich

0.88

 Athletics

0.88

 Dynamo

0.87

 Derby

0.87

Activations Density 0.159%

names of sports teams or clubs

names of football clubs and teams

No Comments

No Known Activations