INDEX

Explanations

words related to hiking and traveling

oai_token-act-pair · gpt-3.5-turbo

references to different groups of people, particularly those involved in travel and outdoor activities

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

lying

-0.64

eals

-0.64

 Pradesh

-0.63

Aw

-0.62

fman

-0.62

orie

-0.60

lesiastical

-0.60

angan

-0.60

lied

-0.59

 Matters

-0.59

POSITIVE LOGITS

 beware

1.11

who

1.00

hip

0.89

paces

0.88

who

0.83

folk

0.83

 rejoice

0.82

 sidx

0.78

esses

0.77

 Beware

0.77

Activations Density 0.070%

words related to hiking and traveling

references to different groups of people, particularly those involved in travel and outdoor activities

No Comments

No Known Activations