INDEX

Explanations

frequent occurrences of a specific letter or character

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

azi

-0.18

afe

-0.18

chter

-0.17

alt

-0.17

rique

-0.16

Ð¿Ð¾Ñĩ

-0.16

croll

-0.16

ales

-0.16

ubl

-0.16

altung

-0.15

POSITIVE LOGITS

ver

0.20

gren

0.18

est

0.17

lyph

0.16

ret

0.16

ez

0.15

overs

0.15

 Chew

0.15

 detail

0.15

estre

0.14

Activations Density 0.012%

frequent occurrences of a specific letter or character

No Comments

No Known Activations