INDEX

Explanations

words related to data manipulation or extraction actions, like "siphoning", "dredging", "shredding", and "divulging"

oai_token-act-pair · gpt-3.5-turbo

gerunds and present participles, particularly focusing on actions or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Supplement

-0.62

eq

-0.60

mbuds

-0.56

 Hume

-0.53

Cos

-0.52

 Moder

-0.52

chev

-0.51

 arbit

-0.50

rud

-0.50

 Compl

-0.50

POSITIVE LOGITS

eering

0.66

staking

0.64

eers

0.63

querade

0.63

ionage

0.62

ishable

0.60

igate

0.60

ources

0.60

source

0.60

tin

0.59

Activations Density 0.172%

words related to data manipulation or extraction actions, like "siphoning", "dredging", "shredding", and "divulging"

gerunds and present participles, particularly focusing on actions or processes

No Comments

No Known Activations