INDEX

Explanations

words related to conflicts or standoffs

oai_token-act-pair · gpt-3.5-turbo

terms related to violent confrontations or conflicts

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

odor

-0.84

urers

-0.70

uries

-0.69

ringe

-0.68

oral

-0.68

incarn

-0.66

yi

-0.66

rates

-0.66

ools

-0.65

broad

-0.65

POSITIVE LOGITS

 standoff

1.32

 SWAT

0.82

 Bundy

0.72

halla

0.71

 Shooter

0.71

 selfies

0.66

 saga

0.66

roy

0.66

emate

0.66

ishly

0.65

Activations Density 0.033%

words related to conflicts or standoffs

terms related to violent confrontations or conflicts

No Comments

No Known Activations