INDEX

Explanations

quantitative comparisons or quantities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "more" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Prediction

-0.71

 Everywhere

-0.67

 wherever

-0.63

 Thomson

-0.60

uca

-0.60

Eye

-0.60

OX

-0.58

 Evening

-0.58

xtap

-0.57

selves

-0.57

POSITIVE LOGITS

 than

1.39

than

1.13

 mundane

1.02

 stringent

1.02

 recent

0.98

 affluent

0.97

 advanced

0.97

 esoteric

0.94

 sophisticated

0.93

 obscure

0.91

Activations Density 0.092%

quantitative comparisons or quantities

instances of the word "more" across various contexts

No Comments

No Known Activations