INDEX

Explanations

references to a specific television news network (Fox News)

oai_token-act-pair · gpt-3.5-turbo

mentions of the Fox News channel

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acterial

-0.77

ansom

-0.72

akeru

-0.72

rians

-0.71

 inval

-0.69

rian

-0.69

orically

-0.67

abulary

-0.66

 Scotia

-0.66

attled

-0.65

POSITIVE LOGITS

conn

1.47

 News

0.98

News

0.97

woods

0.84

hawk

0.83

Planet

0.82

Fox

0.81

FOX

0.80

Wire

0.79

fire

0.78

Activations Density 0.016%

references to a specific television news network (Fox News)

mentions of the Fox News channel

No Comments

No Known Activations