INDEX

Explanations

phrases related to controversies or political figures

oai_token-act-pair · gpt-3.5-turbo

instances of the word "us."

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

ottest

-0.75

regor

-0.74

rought

-0.67

merce

-0.65

jriwal

-0.64

owler

-0.63

skirts

-0.62

ITNESS

-0.62

attery

-0.61

 payoff

-0.60

POSITIVE LOGITS

peed

1.03

pex

1.01

pecting

0.99

pect

0.97

sein

0.93

pects

0.93

cus

0.89

hee

0.88

aurus

0.86

cules

0.86

Activations Density 0.030%

phrases related to controversies or political figures

instances of the word "us."

No Comments

No Known Activations