INDEX

Explanations

phrases ending in '-ced' that might indicate a decision or action

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ced"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

INGTON

-0.67

è¦ļéĨĴ

-0.66

 expenditure

-0.64

 spending

-0.63

ads

-0.61

net

-0.60

 scanners

-0.57

 advertisements

-0.57

 Department

-0.56

 balloon

-0.56

POSITIVE LOGITS

ced

4.73

cing

2.90

ces

2.48

cedes

1.82

cer

1.68

ce

1.51

cible

1.44

cers

1.44

cest

1.32

ched

1.31

Activations Density 0.011%

phrases ending in '-ced' that might indicate a decision or action

occurrences of the word "ced"

No Comments

No Known Activations