INDEX

Explanations

phrases related to sharp or cutting objects

oai_token-act-pair · gpt-3.5-turbo

negative or critical language related to sharp objects or confrontation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĨĴ

-0.72

 Solitaire

-0.64

 Territories

-0.64

 hospitality

-0.63

 frustrations

-0.63

 spont

-0.60

 Ceres

-0.59

 accommodations

-0.59

 jurisdiction

-0.59

 resid

-0.59

POSITIVE LOGITS

equipped

1.24

carry

1.24

shaped

1.22

clad

1.20

laden

1.18

sized

1.16

length

1.15

ridden

1.13

operated

1.12

headed

1.09

Activations Density 0.056%

phrases related to sharp or cutting objects

negative or critical language related to sharp objects or confrontation

No Comments

No Known Activations