INDEX

Explanations

expressions related to personal experiences and achievements

oai_token-act-pair · gpt-3.5-turbo

expressions of gratitude and appreciation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

asury

-0.60

 Officials

-0.60

aspers

-0.60

bsp

-0.60

 Submit

-0.59

ãĥĥãĥĪ

-0.58

ãĥķãĤ¡

-0.57

umo

-0.56

PIN

-0.56

ãĥĥãĤ¯

-0.55

POSITIVE LOGITS

 sucks

1.39

 helps

1.38

 reminds

1.36

 means

1.36

 wasn

1.33

 makes

1.29

 entails

1.29

 inspires

1.28

 meant

1.28

 gives

1.28

Activations Density 0.442%

expressions related to personal experiences and achievements

expressions of gratitude and appreciation

No Comments

No Known Activations