INDEX

Explanations

references to locations or entities related to Los Angeles

oai_token-act-pair · gpt-3.5-turbo

mentions of "LA" (Los Angeles) in various 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

lers

-0.87

manship

-0.82

lication

-0.77

ler

-0.74

ullah

-0.73

theless

-0.71

addons

-0.69

cliffe

-0.68

é¾įå¥ĳå£«

-0.67

folios

-0.67

POSITIVE LOGITS

UNCH

1.24

 Angeles

0.99

UGH

0.94

URA

0.92

KE

0.90

 Galaxy

0.80

 Lakers

0.79

RB

0.78

 Lumpur

0.76

uate

0.76

Activations Density 0.009%

references to locations or entities related to Los Angeles

mentions of "LA" (Los Angeles) in various contexts

No Comments

No Known Activations