INDEX

Explanations

references to the cost of different items or services

oai_token-act-pair · gpt-3.5-turbo

references to the concept of cost

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

ĸļ

-0.94

istar

-0.78

wcs

-0.77

Nap

-0.76

ittees

-0.76

wcsstore

-0.75

ï¸ı

-0.75

AppData

-0.74

rev

-0.73

rist

-0.72

POSITIVE LOGITS

 goods

0.81

 each

0.78

paying

0.73

 groceries

0.73

 maintaining

0.69

 untreated

0.69

 renting

0.69

keeping

0.69

 commuting

0.68

 borrowing

0.68

Activations Density 0.090%

references to the cost of different items or services

references to the concept of cost

No Comments

No Known Activations