INDEX

Explanations

instances of the letter 'h' in various formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

pcs

-0.15

andles

-0.15

ours

-0.15

ats

-0.14

enz

-0.14

WN

-0.14

utas

-0.14

 deficient

-0.14

orns

-0.13

Ð¾Ð»ÐµÑĤ

-0.13

POSITIVE LOGITS

iper

0.17

Ø¯Ùĩ

0.15

alog

0.15

ister

0.15

Ã´te

0.15

obb

0.14

Ĺ

0.14

lm

0.14

(ic

0.14

agi

0.14

Activations Density 0.011%

instances of the letter 'h' in various formats

No Comments

No Known Activations