INDEX

Explanations

references to transportation payment systems

oai_token-act-pair · gpt-3.5-turbo

mentions of various types of "fare" in different contexts

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

icus

-0.86

ieth

-0.83

ultan

-0.81

icio

-0.77

acca

-0.75

antha

-0.73

izations

-0.72

ulkan

-0.71

iversal

-0.70

Klu

-0.69

POSITIVE LOGITS

 fare

1.07

 fares

1.01

fare

1.00

ttes

0.92

well

0.85

bill

0.80

ways

0.79

 Fare

0.77

ptin

0.77

cloth

0.74

Activations Density 0.012%

references to transportation payment systems

mentions of various types of "fare" in different contexts

No Comments

No Known Activations