INDEX

Explanations

proper nouns related to the name "Roberts"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Roberts" in relation to legal or eligibility issues

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

liest

-0.80

alities

-0.69

cond

-0.66

ascar

-0.66

itably

-0.64

bernatorial

-0.63

mble

-0.63

 experien

-0.62

Spread

-0.62

pron

-0.62

POSITIVE LOGITS

haw

1.17

eln

0.96

ullivan

0.87

burgh

0.87

otle

0.83

 Roberts

0.82

rand

0.81

burg

0.81

sey

0.79

icum

0.79

Activations Density 0.012%

proper nouns related to the name "Roberts"

occurrences of the name "Roberts" in relation to legal or eligibility issues

No Comments

No Known Activations