INDEX

Explanations

names or terms related to people, places, and concepts with 'en' at the end

oai_token-act-pair · gpt-3.5-turbo

instances of the suffix "en" in words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 hitch

-0.79

 reconc

-0.68

 shake

-0.67

TPPStreamerBot

-0.67

 Nicarag

-0.65

milo

-0.64

 Canaveral

-0.63

mson

-0.62

 reluct

-0.62

enhagen

-0.61

POSITIVE LOGITS

emies

1.27

chant

1.14

riched

1.08

zyme

1.06

wald

1.05

ema

1.01

ews

0.99

esis

0.99

emy

0.98

viron

0.98

Activations Density 0.034%

names or terms related to people, places, and concepts with 'en' at the end

instances of the suffix "en" in words

No Comments

No Known Activations