INDEX

Explanations

words connected to sorting and categorization based on specific criteria

oai_token-act-pair · gpt-3.5-turbo

phrases related to categorization or organization methods

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hai

-0.77

sil

-0.76

hend

-0.75

Guard

-0.71

bj

-0.70

ibi

-0.70

kee

-0.68

bur

-0.68

assisted

-0.67

?????

-0.67

POSITIVE LOGITS

 geography

1.14

 severity

1.09

 attractiveness

1.04

 geographic

0.99

 demographics

0.96

 geographical

0.96

 latitude

0.96

 ethnicity

0.95

 gender

0.94

 nationality

0.92

Activations Density 0.257%

words connected to sorting and categorization based on specific criteria

phrases related to categorization or organization methods

No Comments

No Known Activations