INDEX

Explanations

events or actions happening in a sequence

oai_token-act-pair · gpt-3.5-turbo

instances of the word "when" indicating temporal sequences or events

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

zag

-0.82

Í

-0.75

ouble

-0.72

species

-0.71

cheat

-0.71

JV

-0.71

whatever

-0.70

igmatic

-0.70

abal

-0.69

tick

-0.68

POSITIVE LOGITS

 asked

1.49

 confronted

1.42

 questioned

1.34

 contacted

1.33

 pressed

1.28

 approached

1.15

 quizz

1.11

soever

1.04

 interviewed

1.03

 challenged

1.03

Activations Density 0.108%

events or actions happening in a sequence

instances of the word "when" indicating temporal sequences or events

No Comments

No Known Activations