INDEX

Explanations

terms related to legal districts and locations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific geographical districts, particularly in legal or administrative contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ppo

-0.87

cial

-0.86

issan

-0.85

ggie

-0.80

gyn

-0.78

ggy

-0.78

ggies

-0.78

ctors

-0.77

gger

-0.74

cius

-0.74

POSITIVE LOGITS

 District

1.01

District

0.88

rict

0.81

Apps

0.78

 Attorney

0.73

reth

0.72

 Passage

0.66

mite

0.66

 district

0.65

 Marshal

0.65

Activations Density 0.011%

terms related to legal districts and locations

mentions of specific geographical districts, particularly in legal or administrative contexts

No Comments

No Known Activations