INDEX

Explanations

mentions of conflict zones and military activities

oai_token-act-pair · gpt-3.5-turbo

references to rebel-held areas and their conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Mercury

-1.01

 Wilde

-0.91

Scot

-0.87

 Einstein

-0.82

Consumer

-0.82

 Scotch

-0.78

çĶ°

-0.78

 Freem

-0.77

aughs

-0.77

 Titanic

-0.75

POSITIVE LOGITS

 besieged

0.98

merga

0.97

orsi

0.96

 ceasefire

0.95

 rebels

0.92

 rebel

0.90

 militias

0.89

 militia

0.88

 Idlib

0.87

 insurgents

0.87

Activations Density 0.291%

mentions of conflict zones and military activities

references to rebel-held areas and their conditions

No Comments

No Known Activations