INDEX

Explanations

phrases related to continuous or ongoing action

oai_token-act-pair · gpt-3.5-turbo

the word "doing" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

lights

-0.69

Tier

-0.69

mare

-0.67

mares

-0.66

)=(

-0.66

flags

-0.65

liner

-0.65

pter

-0.64

case

-0.63

sg

-0.63

POSITIVE LOGITS

berman

0.80

omething

0.79

pez

0.77

 nothing

0.76

 something

0.74

ggy

0.73

VIDIA

0.73

å¥³

0.73

 terribly

0.71

lyak

0.71

Activations Density 0.036%

phrases related to continuous or ongoing action

the word "doing" in various contexts

No Comments

No Known Activations