INDEX

Explanations

phrases related to percentages and proportions

oai_token-act-pair · gpt-3.5-turbo

repeated instances of the word "of" in relation to percentages, measures, or proportions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

mpeg

-0.68

hunt

-0.66

andal

-0.64

pedia

-0.63

æĦ

-0.59

sort

-0.58

tests

-0.57

consider

-0.57

abl

-0.57

alph

-0.57

POSITIVE LOGITS

GDP

1.07

GN

0.81

 total

0.74

 households

0.69

 humanity

0.66

 respondents

0.65

 earners

0.63

its

0.63

 overall

0.61

 their

0.60

Activations Density 0.082%

phrases related to percentages and proportions

repeated instances of the word "of" in relation to percentages, measures, or proportions

No Comments

No Known Activations