INDEX

Explanations

punctuation marks, especially closing brackets and parentheses

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

↵

-0.15

baugh

-0.15

idable

-0.14

Ã©ry

-0.14

asons

-0.14

ie

-0.14

afs

-0.13

stringLiteral

-0.13

POSITIVE LOGITS

yon

0.17

jpg

0.17

zano

0.15

ÙĨØª

0.15

riet

0.15

toi

0.14

affer

0.14

.↵↵↵↵↵↵

0.14

kest

0.14

Activations Density 0.042%

punctuation marks, especially closing brackets and parentheses

No Known Activations