INDEX

Explanations

the surname "son"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the suffix "son" in names or terms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 heels

-0.76

ANN

-0.72

 neut

-0.64

 breaches

-0.64

pse

-0.63

 spaces

-0.63

 deficits

-0.63

 chronically

-0.63

 benign

-0.62

 contagious

-0.62

POSITIVE LOGITS

son

0.98

ville

0.94

nen

0.93

strom

0.91

ning

0.88

nets

0.87

fman

0.85

hood

0.85

racuse

0.84

emouth

0.83

Activations Density 0.011%

the surname "son"

occurrences of the suffix "son" in names or terms

No Comments

No Known Activations