INDEX

Explanations

specific nouns followed by possessive 's

oai_token-act-pair · gpt-3.5-turbo

words related to legal terminology and concepts

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

ones

-0.72

 Bundy

-0.62

Fight

-0.61

 Topic

-0.61

INTON

-0.60

stad

-0.60

 Serving

-0.60

someone

-0.59

Unknown

-0.59

Reviewed

-0.58

POSITIVE LOGITS

 woes

0.85

ãĥķãĤ©

0.80

 spree

0.72

 selves

0.67

 demise

0.65

anmar

0.65

 antics

0.64

 territories

0.64

 takeover

0.63

 rampage

0.62

Activations Density 0.331%

specific nouns followed by possessive 's

words related to legal terminology and concepts

No Comments

No Known Activations