INDEX

Explanations

terms related to sports offenses

oai_token-act-pair · gpt-3.5-turbo

references to offensive strategies in sports

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

bourg

-0.81

EVA

-0.78

azo

-0.71

ioxide

-0.67

AFTA

-0.67

grave

-0.66

leaf

-0.66

obook

-0.65

éĸ

-0.65

 Jenner

-0.64

POSITIVE LOGITS

 Offense

0.92

 offense

0.92

ENSE

0.74

 oriented

0.73

 Rodgers

0.73

 linemen

0.73

scill

0.72

Cre

0.69

 offensively

0.67

 lineman

0.66

Activations Density 0.019%

terms related to sports offenses

references to offensive strategies in sports

No Comments

No Known Activations