INDEX

Explanations

information related to quantities or values, such as numbers and measurements

oai_token-act-pair · gpt-3.5-turbo

statements about statistics or measurements related to financial or demographic data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

wake

-0.68

 Awakens

-0.65

attach

-0.64

ossom

-0.62

 wills

-0.62

 refrain

-0.62

practice

-0.62

stop

-0.62

 Offer

-0.59

 Particip

-0.59

POSITIVE LOGITS

 negligible

1.14

 estimated

1.14

 pegged

1.02

 higher

1.01

 considerably

1.00

 reduced

1.00

 exceeded

0.99

 roughly

0.97

 greater

0.96

 tripled

0.95

Activations Density 0.142%

information related to quantities or values, such as numbers and measurements

statements about statistics or measurements related to financial or demographic data

No Comments

No Known Activations