INDEX

Explanations

instances of the word "Ghomi"

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the word "hi."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

lain

-0.80

nesday

-0.77

swick

-0.71

argon

-0.71

ividual

-0.68

orative

-0.68

tle

-0.68

ktop

-0.66

rations

-0.66

atility

-0.64

POSITIVE LOGITS

emi

1.02

ya

1.01

yy

0.92

kson

0.88

yah

0.87

pper

0.87

veland

0.86

ELD

0.84

qq

0.84

pping

0.83

Activations Density 0.019%

instances of the word "Ghomi"

repeated mentions of the word "hi."

No Comments

No Known Activations