INDEX

Explanations

words related to non-conformity and non-traditional forms

oai_token-act-pair · gpt-3.5-turbo

references to nonconformity or alternative formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Liqu

-0.62

 Royals

-0.62

ILA

-0.60

 paras

-0.59

mine

-0.58

mins

-0.58

 Ammunition

-0.57

 Breath

-0.57

nea

-0.56

 Regional

-0.56

POSITIVE LOGITS

idable

1.31

aldehyde

1.25

atted

1.23

ulations

1.07

ative

1.07

ula

1.05

ational

1.03

ulation

0.93

ulate

0.91

ulator

0.88

Activations Density 0.017%

words related to non-conformity and non-traditional forms

references to nonconformity or alternative formats

No Comments

No Known Activations