INDEX

Explanations

instances of the word "agen" in the text

oai_token-act-pair · gpt-3.5-turbo

the term "agen" 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 @ 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

ODUCT

-0.67

 tradem

-0.65

Tec

-0.63

 Investigator

-0.62

 Reese

-0.60

WHERE

-0.60

TI

-0.59

MQ

-0.58

 cytok

-0.57

SIG

-0.57

POSITIVE LOGITS

furt

1.53

esis

1.40

etic

1.08

arian

1.04

etics

1.02

arians

0.97

hin

0.92

raham

0.88

heim

0.87

arde

0.85

Activations Density 0.005%

instances of the word "agen" in the text

the term "agen" in various contexts

No Comments

No Known Activations