INDEX

Explanations

questions or unclear statements

oai_token-act-pair · gpt-3.5-turbo

questions and expressions of uncertainty or confusion

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

opsis

-0.67

aldi

-0.66

 marrow

-0.65

arers

-0.64

dy

-0.62

 microbiome

-0.62

earable

-0.61

assium

-0.61

 helicop

-0.61

 handshake

-0.61

POSITIVE LOGITS

ï¿½ï¿½ï¿½ï¿½

1.01

?,

0.91

.?

0.91

???

0.89

STER

0.82

 Nope

0.75

?:

0.75

Huh

0.75

??

0.73

furt

0.73

Activations Density 0.035%

questions or unclear statements

questions and expressions of uncertainty or confusion

No Comments

No Known Activations