INDEX

Explanations

terms related to specific names, events, and organizations

oai_token-act-pair · gpt-3.5-turbo

names of organizations, individuals, and specific legislations or events

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

Reviewer

-0.58

requires

-0.51

 Soph

-0.48

 Corrections

-0.48

ear

-0.47

hiro

-0.45

hand

-0.45

Pub

-0.45

pub

-0.45

tips

-0.45

POSITIVE LOGITS

 TAMADRA

0.52

 Jinping

0.50

 successors

0.49

 Skydragon

0.49

quished

0.48

·

0.47

ADRA

0.47

(=

0.47

èĢħ

0.46

 adherents

0.45

Activations Density 0.524%

terms related to specific names, events, and organizations

names of organizations, individuals, and specific legislations or events

No Comments

No Known Activations