INDEX

Explanations

past tense verbs indicating actions taken by someone or something

oai_token-act-pair · gpt-3.5-turbo

the conditional or future tense used in sentences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

values

-0.75

resources

-0.67

rylic

-0.64

holder

-0.63

meta

-0.62

building

-0.62

è¦ļéĨĴ

-0.62

 Efficiency

-0.62

advant

-0.61

blocking

-0.60

POSITIVE LOGITS

be

1.04

ĸļ

0.99

 doubtless

0.97

 undoubtedly

0.93

 surely

0.91

 inevitably

0.90

 suffice

0.89

 eventually

0.88

 never

0.87

 dearly

0.83

Activations Density 0.161%

past tense verbs indicating actions taken by someone or something

the conditional or future tense used in sentences

No Comments

No Known Activations