INDEX

Explanations

mentions of natural disasters, specifically wildfires

oai_token-act-pair · gpt-3.5-turbo

references to wildfires or fire incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Birth

-0.75

Tok

-0.72

 Barcl

-0.66

 Surgery

-0.65

WOR

-0.63

afort

-0.62

çİĭ

-0.61

GB

-0.61

conservative

-0.61

Students

-0.61

POSITIVE LOGITS

hooting

0.94

 fires

0.92

fires

0.90

 torches

0.88

paces

0.86

linger

0.85

 Fired

0.84

 flares

0.83

 blazing

0.82

 retard

0.81

Activations Density 0.007%

mentions of natural disasters, specifically wildfires

references to wildfires or fire incidents

No Comments

No Known Activations