INDEX

Explanations

references to explosives

oai_token-act-pair · gpt-3.5-turbo

references to explosives and related terms

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

aird

-0.89

 Seah

-0.76

)=(

-0.73

naire

-0.72

esan

-0.71

 Goose

-0.69

ups

-0.69

heit

-0.69

gling

-0.67

SEA

-0.66

POSITIVE LOGITS

 detonated

1.10

 deton

0.98

 barric

0.92

 explosives

0.91

 deterrent

0.89

 disposal

0.86

 detectors

0.83

 detector

0.81

ilitary

0.79

kit

0.78

Activations Density 0.029%

references to explosives

references to explosives and related terms

No Comments

No Known Activations