INDEX

Explanations

references to armed groups or fighters, particularly militants

oai_token-act-pair · gpt-3.5-turbo

mentions of militant groups and related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

IENCE

-0.80

iture

-0.77

Sense

-0.76

âĶĢâĶĢâĶĢâĶĢ

-0.74

 Goose

-0.71

ĳ

-0.71

DATA

-0.70

ORK

-0.69

ðĿ

-0.69

othy

-0.69

POSITIVE LOGITS

 insurgents

1.06

 gunmen

1.04

 insurgency

1.00

 infiltr

1.00

 affiliated

0.98

 besie

0.97

 militants

0.96

 terrorists

0.93

 occupying

0.93

 Islamists

0.92

Activations Density 0.072%

references to armed groups or fighters, particularly militants

mentions of militant groups and related terminology

No Comments

No Known Activations