INDEX

Explanations

proper nouns related to specific people, places, or events

oai_token-act-pair · gpt-3.5-turbo

specific names, terms, and entities, particularly in a contextual or historical framework

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

-0.87

 sqor

-0.80

 guiActiveUnfocused

-0.60

,,,,

-0.59

[&

-0.58

jun

-0.57

izabeth

-0.57

 istg

-0.57

stadt

-0.57

kees

-0.56

POSITIVE LOGITS

FAQ

0.64

APD

0.63

boarding

0.59

istar

0.59

eral

0.58

Net

0.56

COR

0.55

 Telecom

0.55

ong

0.54

iots

0.54

Activations Density 0.454%

proper nouns related to specific people, places, or events

specific names, terms, and entities, particularly in a contextual or historical framework

No Comments

No Known Activations