INDEX

Explanations

instances of the word "this" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

orman

-0.16

nt

-0.16

etc

-0.15

 Gates

-0.15

elson

-0.15

oubted

-0.14

odd

-0.14

fails

-0.14

oney

-0.14

wiki

-0.14

POSITIVE LOGITS

is

0.23

 ones

0.18

 whole

0.18

æĺ¯æĪĳ

0.18

Whole

0.18

 entire

0.17

ãģ¯

0.16

isn

0.15

âĢĮØ³

0.15

Activations Density 0.133%

instances of the word "this" in various forms

No Comments

No Known Activations