INDEX

Explanations

the word "Sh" followed by different suffixes or in combination with other letters

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Sh" or variations thereof in the text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

etheless

-0.96

anwhile

-0.87

ãĥ¼ãĥĨãĤ£

-0.75

 confir

-0.73

 Ajax

-0.73

 numbered

-0.69

ãĥ¼ãĥĨ

-0.69

OLOGY

-0.69

 Corpus

-0.66

 foil

-0.65

POSITIVE LOGITS

aky

1.10

ooting

1.09

allow

1.09

isha

1.07

atters

1.06

attered

1.06

uff

1.06

attering

1.06

osh

1.05

ifty

1.04

Activations Density 0.021%

the word "Sh" followed by different suffixes or in combination with other letters

mentions of the name "Sh" or variations thereof in the text

No Comments

No Known Activations