INDEX

Explanations

instructions or recommendations to perform certain actions

oai_token-act-pair · gpt-3.5-turbo

suggestions or recommendations for actions

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

hiba

-0.69

lish

-0.65

gren

-0.62

meta

-0.62

ophone

-0.62

 glitch

-0.61

soDeliveryDate

-0.61

wikipedia

-0.60

interstitial

-0.58

VERTISEMENT

-0.58

POSITIVE LOGITS

reprene

0.84

urities

0.70

ate

0.68

anamo

0.67

ter

0.67

ãĤ¦ãĤ¹

0.65

 warr

0.60

htar

0.59

to

0.58

BAT

0.58

Activations Density 0.056%

instructions or recommendations to perform certain actions

suggestions or recommendations for actions

No Comments

No Known Activations