INDEX

Explanations

numbers indicating specific quantities or measurements

oai_token-act-pair · gpt-3.5-turbo

specific numerical values or identifiers

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

arette

-0.79

iful

-0.75

asio

-0.74

jad

-0.73

ribune

-0.72

hesis

-0.71

actic

-0.70

heet

-0.70

oken

-0.69

arettes

-0.69

POSITIVE LOGITS

 ILCS

0.78

emale

0.77

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.75

lbs

0.75

rf

0.75

RF

0.74

mph

0.72

LR

0.65

ĺħ

0.64

itably

0.64

Activations Density 0.032%

numbers indicating specific quantities or measurements

specific numerical values or identifiers

No Comments

No Known Activations