INDEX

Explanations

mentions of sports teams, especially the Chicago Bears, and their games and rivalries

oai_token-act-pair · gpt-3.5-turbo

mentions of the Chicago Bears football team

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

akuya

-0.80

oppable

-0.80

asive

-0.79

ascular

-0.76

ADRA

-0.76

istry

-0.76

employment

-0.73

iltration

-0.72

CLASSIFIED

-0.69

egal

-0.67

POSITIVE LOGITS

 Bears

1.33

 Tribune

0.84

 Drill

0.82

 Bryant

0.81

 Cubs

0.80

heimer

0.80

eman

0.79

fires

0.79

 Packers

0.78

 Grizz

0.76

Activations Density 0.012%

mentions of sports teams, especially the Chicago Bears, and their games and rivalries

mentions of the Chicago Bears football team

No Comments

No Known Activations