INDEX

Explanations

parentheses used to provide additional information or clarification within a sentence

oai_token-act-pair · gpt-3.5-turbo

occurrences of parentheses in the text

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

 flock

-0.77

dow

-0.74

 inund

-0.72

 glare

-0.69

 coron

-0.69

 piping

-0.67

 entitle

-0.64

 dazz

-0.63

 tack

-0.62

 inquest

-0.62

POSITIVE LOGITS

sic

1.39

emphasis

1.35

possibly

1.31

albeit

1.24

...)

1.22

again

1.22

âĢ¦)

1.21

?)

1.19

excluding

1.17

perhaps

1.16

Activations Density 0.111%

parentheses used to provide additional information or clarification within a sentence

occurrences of parentheses in the text

No Comments

No Known Activations