INDEX

Explanations

mentions of the news channel "Fox News"

oai_token-act-pair · gpt-3.5-turbo

references to "Fox News" within the text

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

acterial

-0.73

chest

-0.68

paying

-0.68

 Scher

-0.68

ysis

-0.65

isu

-0.64

ript

-0.64

cy

-0.64

 sheet

-0.63

ritis

-0.63

POSITIVE LOGITS

 Channel

1.17

 Networks

0.82

Radio

0.80

Channel

0.80

 Interactive

0.80

 Conference

0.77

 Radio

0.77

 Network

0.77

HQ

0.76

 Headquarters

0.76

Activations Density 0.026%

mentions of the news channel "Fox News"

references to "Fox News" within the text

No Comments

No Known Activations