INDEX

Explanations

legal case names

oai_token-act-pair · gpt-3.5-turbo

legal cases and references to court decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

inki

-0.84

pecially

-0.84

 enthusi

-0.71

pport

-0.71

ahime

-0.70

bley

-0.69

bender

-0.69

onga

-0.69

swick

-0.68

oji

-0.67

POSITIVE LOGITS

 supra

0.93

 Commonwealth

0.78

 Kemp

0.72

 Ridge

0.69

 Olympia

0.68

 Lawson

0.66

 Lauderdale

0.65

 Bowen

0.65

 Koch

0.64

 Kessler

0.64

Activations Density 0.053%

legal case names

legal cases and references to court decisions

No Comments

No Known Activations