INDEX

Explanations

language related to political commentary and statements

oai_token-act-pair · gpt-3.5-turbo

statements and discussions related to political accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

nown

-0.82

natureconservancy

-0.78

MPG

-0.76

 Trails

-0.76

 Navigation

-0.76

 Transition

-0.75

glers

-0.73

maps

-0.71

 datas

-0.70

 feasibility

-0.69

POSITIVE LOGITS

 condemnation

1.67

 rebuke

1.64

 condemn

1.57

 indignation

1.56

 scorn

1.51

 outrage

1.45

 outraged

1.44

 condem

1.43

 condemning

1.42

 denounce

1.42

Activations Density 0.767%

language related to political commentary and statements

statements and discussions related to political accountability

No Comments

No Known Activations