INDEX

Explanations

phrases related to quantifying or specifying a portion or percentage of something

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate percentages or ratios related to various topics or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

fml

-0.75

Lab

-0.68

ASY

-0.66

Attempts

-0.66

athy

-0.62

intend

-0.62

iw

-0.60

PI

-0.60

Whe

-0.59

Truth

-0.58

POSITIVE LOGITS

 roughly

1.05

 nearly

1.01

 approximately

0.97

 majority

0.96

 most

0.95

 brunt

0.94

 âĪ¼

0.92

0.89

 almost

0.87

Activations Density 0.129%

phrases related to quantifying or specifying a portion or percentage of something

phrases that indicate percentages or ratios related to various topics or groups

No Comments

No Known Activations