INDEX

Explanations

references to political regimes and their associated actions

oai_token-act-pair · gpt-3.5-turbo

mentions of governmental or political regimes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

furt

-0.77

sbm

-0.74

TH

-0.74

ths

-0.74

Redd

-0.74

 Lent

-0.73

Kent

-0.72

draft

-0.71

lder

-0.71

 Scotch

-0.71

POSITIVE LOGITS

 regime

1.19

 regimes

1.15

 unilaterally

0.87

 dictator

0.85

 dictatorship

0.85

 imposed

0.82

 governing

0.81

 overthrow

0.80

 overth

0.79

 eleph

0.79

Activations Density 0.007%

references to political regimes and their associated actions

mentions of governmental or political regimes

No Comments

No Known Activations