INDEX

Explanations

mentions of the geographical location "Utah"

oai_token-act-pair · gpt-3.5-turbo

mentions of the state of Utah

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

lihood

-0.79

hops

-0.74

 depreciation

-0.73

enegger

-0.72

acters

-0.69

earance

-0.69

ospace

-0.69

emis

-0.67

 behav

-0.67

alez

-0.66

POSITIVE LOGITS

ns

1.09

 Jazz

1.01

Utah

0.90

 Rapids

0.88

 Territory

0.82

 Valley

0.82

 Yard

0.82

 Utah

0.81

bard

0.79

Policy

0.76

Activations Density 0.008%

mentions of the geographical location "Utah"

mentions of the state of Utah

No Comments

No Known Activations