INDEX

Explanations

instances of the word "round" mentioned in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to rounds or sequential events in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

yer

-0.65

 mathemat

-0.63

ettel

-0.62

KI

-0.61

PLA

-0.61

"},"

-0.60

uras

-0.59

ORY

-0.58

 Bott

-0.57

é¾įå

-0.56

POSITIVE LOGITS

trip

0.93

table

0.92

abouts

0.91

about

0.87

robe

0.86

scribed

0.79

bag

0.77

 Round

0.75

 round

0.75

Round

0.74

Activations Density 0.014%

instances of the word "round" mentioned in various contexts

references to rounds or sequential events in various contexts

No Comments

No Known Activations