INDEX

Explanations

phrases related to multiple entities or items being considered or discussed together

oai_token-act-pair · gpt-3.5-turbo

repeated phrases or connections incorporating the word "and."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

icken

-0.65

inational

-0.60

 guiActiveUnfocused

-0.60

idia

-0.58

idel

-0.57

inates

-0.57

aeus

-0.57

ettel

-0.57

olation

-0.56

schild

-0.55

POSITIVE LOGITS

 others

1.61

 other

1.43

 countless

1.39

 many

1.28

 innumerable

1.13

other

1.09

 dozens

1.09

 Others

1.05

 myriad

1.04

 similar

1.02

Activations Density 0.346%

phrases related to multiple entities or items being considered or discussed together

repeated phrases or connections incorporating the word "and."

No Comments

No Known Activations