INDEX

Explanations

elements related to names and titles of companies, stories, or projects

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

andel

-0.16

pector

-0.15

uka

-0.15

ende

-0.15

hle

-0.14

elon

-0.14

 TYPO

-0.14

Ð¾Ð¼Ñĸ

-0.14

izik

-0.13

æĴĥ

-0.13

POSITIVE LOGITS

geber

0.16

0.15

ialized

0.15

Stamp

0.14

ously

0.14

Gi

0.14

boro

0.14

OfFile

0.14

Adj

0.14

 Establishment

0.14

Activations Density 0.169%

elements related to names and titles of companies, stories, or projects

No Comments

No Known Activations