INDEX

Explanations

capitalized acronyms with 'UT'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the token "UT"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

united

-0.73

perse

-0.70

boards

-0.69

acters

-0.69

compan

-0.68

ãĥ£

-0.65

views

-0.64

strength

-0.64

clad

-0.63

isma

-0.62

POSITIVE LOGITS

TING

1.30

ierrez

0.99

LEY

0.95

TER

0.93

UAL

0.92

UT

0.91

ypes

0.91

UES

0.90

IL

0.89

BLIC

0.87

Activations Density 0.004%

capitalized acronyms with 'UT'

occurrences of the token "UT"

No Comments

No Known Activations