INDEX

Explanations

specific mentions of the word "Court."

oai_token-act-pair · gpt-3.5-turbo

mentions of courts or related legal terminology

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

 rooting

-0.78

 culturally

-0.61

lihood

-0.60

 Horus

-0.59

 curfew

-0.59

 donor

-0.59

 goodbye

-0.59

 economically

-0.58

hol

-0.58

 flashlight

-0.58

POSITIVE LOGITS

tyard

1.50

thouse

1.35

ses

1.12

ser

1.10

cour

1.06

sers

1.06

 Cour

1.04

eneg

0.97

icol

0.96

cil

0.93

Activations Density 0.010%

specific mentions of the word "Court."

mentions of courts or related legal terminology

No Comments

No Known Activations