INDEX

Explanations

names of specific entities or locations

oai_token-act-pair · gpt-3.5-turbo

references to specific organizations, institutions, or geographical locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 councillors

-0.51

 princ

-0.49

ins

-0.49

and

-0.49

arts

-0.48

isations

-0.48

 capitals

-0.48

 seiz

-0.47

eks

-0.47

 organising

-0.47

POSITIVE LOGITS

ļéĨĴ

0.62

Interstitial

0.59

ravity

0.58

edIn

0.54

ãĥĺãĥ©

0.54

Cooldown

0.53

inventoryQuantity

0.51

quished

0.51

ä¹ĭ

0.51

Materials

0.50

Activations Density 1.385%

names of specific entities or locations

references to specific organizations, institutions, or geographical locations

No Comments

No Known Activations