INDEX

Explanations

sections and headings within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

yw

-0.16

 unless

-0.14

plaintext

-0.13

Votes

-0.13

imity

-0.13

quential

-0.13

adu

-0.13

 Ø§ÙĦØ¹ÙĨ

-0.13

VM

-0.13

POSITIVE LOGITS

ghan

0.16

anje

0.16

bakan

0.15

Ã´me

0.15

á»Ļc

0.14

LBL

0.13

Å¡ov

0.13

µ¬

0.13

å¹ķ

0.13

ìĿ´ìĬ¤

0.13

Activations Density 0.085%

sections and headings within text

No Comments

No Known Activations