INDEX

Explanations

legal and criminal terminology

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

terms related to legal charges and offenses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

tnc

-0.75

soDeliveryDate

-0.67

 Runes

-0.61

pron

-0.60

oulos

-0.60

iors

-0.59

 Slack

-0.57

iov

-0.57

osaurus

-0.56

iji

-0.55

POSITIVE LOGITS

 punishable

1.01

 stemming

0.96

 relating

0.92

 apiece

0.89

 count

0.88

 alleging

0.87

 aggravated

0.86

 counts

0.86

 misdem

0.86

 felony

0.86

Activations Density 0.066%

legal and criminal terminology

terms related to legal charges and offenses

No Comments

No Known Activations