INDEX

Explanations

the occurrence of closing tags or statements in a structured document format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

able

-0.15

ito

-0.14

itness

-0.14

lien

-0.14

atomy

-0.14

æ¢°

-0.14

burg

-0.14

 hÃłng

-0.13

 Loren

-0.13

venir

-0.13

POSITIVE LOGITS

icot

0.16

/OR

0.15

urances

0.15

 Bates

0.15

aÄį

0.14

aminer

0.14

/disable

0.14

_LT

0.14

_GB

0.14

ratulations

0.13

Activations Density 0.008%

the occurrence of closing tags or statements in a structured document format

No Comments

No Known Activations