INDEX

Explanations

instances of the word "one" in various forms and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

relude

-0.17

 greater

-0.17

greater

-0.16

llx

-0.16

_greater

-0.16

ouce

-0.15

Greater

-0.15

osci

-0.15

alyze

-0.15

POSITIVE LOGITS

one

0.23

of

0.23

 among

0.20

 Ð¸Ð·

0.19

 salah

0.18

of

0.17

one

0.16

 amongst

0.16

 half

0.16

Of

0.16

Activations Density 0.050%

instances of the word "one" in various forms and contexts

No Comments

No Known Activations