INDEX

Explanations

HTML elements and links within the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

Yun

-0.16

loquent

-0.15

SF

-0.14

inya

-0.14

 misd

-0.14

Ø«ÙĬØ±

-0.14

hai

-0.14

ourd

-0.13

Tor

-0.13

ilm

-0.13

POSITIVE LOGITS

ed

0.17

dy

0.17

idia

0.15

æ¾¤

0.15

Dy

0.14

uer

0.13

edx

0.13

iles

0.13

meer

0.13

Zi

0.13

Activations Density 0.050%

HTML elements and links within the document

No Comments

No Known Activations