INDEX

Explanations

political language and terms

oai_token-act-pair · gpt-3.5-turbo

content related to political positions and commitments

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tics

-0.87

mol

-0.82

zsche

-0.75

zig

-0.74

ilial

-0.73

pmwiki

-0.71

 indic

-0.66

 converter

-0.65

 bondage

-0.65

condition

-0.65

POSITIVE LOGITS

Senator

1.44

Senate

1.44

 Congressman

1.42

 Senator

1.41

 Senate

1.40

 senate

1.40

 senator

1.37

 senators

1.35

 congressman

1.33

 presidential

1.31

Activations Density 1.874%

political language and terms

content related to political positions and commitments

No Comments

No Known Activations