INDEX

Explanations

references to scoring in sports

oai_token-act-pair · gpt-3.5-turbo

references to scoring in a competitive context, particularly in sports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ship

-0.71

ships

-0.69

tan

-0.68

UAL

-0.68

fer

-0.68

lly

-0.66

agan

-0.64

conn

-0.63

san

-0.63

tor

-0.62

POSITIVE LOGITS

 scoring

0.99

otle

0.85

 scored

0.81

 scorer

0.78

ificant

0.77

cards

0.76

 enthusi

0.76

 score

0.74

scoring

0.73

lass

0.73

Activations Density 0.006%

references to scoring in sports

references to scoring in a competitive context, particularly in sports

No Comments

No Known Activations