INDEX

Explanations

organizations and entities related to the media and politics

oai_token-act-pair · gpt-3.5-turbo

references to the media and its relationship with public perception and trust

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

edIn

-0.84

Ô

-0.82

conserv

-0.75

perse

-0.74

ĪĴ

-0.73

ļé

-0.73

ãĤ¨ãĥ«

-0.71

¶æ

-0.70

chwitz

-0.70

ioned

-0.69

POSITIVE LOGITS

 itself

0.97

 microbiome

0.82

 hierarchy

0.78

ousel

0.77

 Gazette

0.71

liest

0.69

 chorus

0.69

 floor

0.69

's

0.69

venth

0.67

Activations Density 0.524%

organizations and entities related to the media and politics

references to the media and its relationship with public perception and trust

No Comments

No Known Activations