INDEX

Explanations

words related to absolute actions or complete states

oai_token-act-pair · gpt-3.5-turbo

instances of the word "all" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

•All the Features!

No Comments

Negative Logits

yip

-0.65

SHIP

-0.64

gee

-0.63

raid

-0.63

lav

-0.63

olate

-0.63

icion

-0.62

potion

-0.61

 Massive

-0.60

Led

-0.59

POSITIVE LOGITS

 traces

1.04

usions

1.03

 kinds

0.97

ude

0.95

 sorts

0.90

 semblance

0.89

uding

0.89

 else

0.88

udes

0.86

igators

0.86

Activations Density 0.070%

words related to absolute actions or complete states

instances of the word "all" and its variations in different contexts

No Comments

No Known Activations