INDEX

Explanations

numbers and quantities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "in."

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

DAQ

-0.71

igers

-0.64

ife

-0.63

 Quan

-0.61

anamo

-0.60

SPA

-0.58

 Panda

-0.56

Sharp

-0.56

Alan

-0.55

istani

-0.55

POSITIVE LOGITS

 fact

0.72

ivari

0.67

 comprom

0.66

 addition

0.64

etheless

0.61

romy

0.61

 contrast

0.61

ptr

0.61

olesc

0.60

yss

0.60

Activations Density 0.633%

numbers and quantities

occurrences of the word "in."

No Comments

No Known Activations