INDEX

Explanations

numerical data and values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

eriod

-0.15

anye

-0.15

erot

-0.15

conto

-0.14

peg

-0.14

spl

-0.14

Spl

-0.13

awan

-0.13

 Percent

-0.13

sumer

-0.13

POSITIVE LOGITS

per

0.94

_per

0.62

per

0.57

-per

0.50

Per

0.46

Per

0.44

_PER

0.42

.per

0.41

 Ð¿ÐµÑĢ

0.37

(per

0.35

Activations Density 0.179%

numerical data and values

No Comments

No Known Activations

numerical data and values

No Comments

No Known Activations