INDEX

Explanations

phrases related to unspecified quantities or entities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "some."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 guide

-0.64

 Kamp

-0.62

 lull

-0.62

 Guide

-0.58

elta

-0.57

 firmly

-0.57

 prim

-0.56

izer

-0.56

 rolling

-0.56

Vie

-0.56

POSITIVE LOGITS

some

4.33

something

1.75

many

1.36

someone

1.27

Some

1.25

mostly

1.18

another

1.16

maybe

1.15

sometimes

1.13

these

1.11

Activations Density 0.011%

phrases related to unspecified quantities or entities

occurrences of the word "some."

No Comments

No Known Activations