INDEX

Explanations

words related to legal proceedings or legal terminology

oai_token-act-pair · gpt-3.5-turbo

references to legal trials

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

flush

-0.82

olen

-0.74

CRE

-0.73

VP

-0.72

DEBUG

-0.68

Tik

-0.64

respect

-0.62

oji

-0.62

jl

-0.61

uts

-0.61

POSITIVE LOGITS

 trial

1.24

 Trial

1.11

 Trials

1.10

 trials

1.08

trial

0.87

 hearings

0.84

neys

0.84

imony

0.82

 "$:/

0.81

suit

0.75

Activations Density 0.010%

words related to legal proceedings or legal terminology

references to legal trials

No Comments

No Known Activations