INDEX

Explanations

proper nouns related to different entities, potentially organizations or individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns and titles, particularly names and significant terms

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

 âĢİ

-0.54

 Lyme

-0.52

 opposition

-0.49

 Transformers

-0.49

 later

-0.48

 PokÃ©mon

-0.46

 grounds

-0.46

Bat

-0.45

 fertile

-0.44

Merit

-0.44

POSITIVE LOGITS

etheless

0.85

anmar

0.72

imil

0.66

etheus

0.66

obin

0.66

milo

0.65

theless

0.65

rontal

0.65

apest

0.64

querade

0.64

Activations Density 0.765%

proper nouns related to different entities, potentially organizations or individuals

proper nouns and titles, particularly names and significant terms

No Comments

No Known Activations