INDEX

Explanations

references to jihadist groups or activities

oai_token-act-pair · gpt-3.5-turbo

terms related to jihadist groups and movements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Veter

-0.77

ances

-0.76

 Celestial

-0.75

ðĿ

-0.74

nih

-0.69

 Matter

-0.67

Rye

-0.67

 Volt

-0.67

 Rept

-0.66

 Neck

-0.66

POSITIVE LOGITS

 jihadists

0.92

 jihadist

0.87

 terrorists

0.84

 jihad

0.83

istan

0.82

 fighters

0.80

 terrorist

0.80

 extremists

0.79

 terror

0.78

ihadi

0.78

Activations Density 0.011%

references to jihadist groups or activities

terms related to jihadist groups and movements

No Comments

No Known Activations