INDEX

Explanations

abbreviated references or initials followed by punctuation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

annel

-0.15

otor

-0.15

 loose

-0.15

Ð¸ÑĨ

-0.14

usch

-0.14

Äįe

-0.14

ANNEL

-0.14

arts

-0.14

 feeling

-0.14

ogne

-0.13

POSITIVE LOGITS

Ð¶Ð´

0.15

æŁ

0.15

Dw

0.15

/Runtime

0.15

Spe

0.14

Ð¶Ð°

0.14

ocalypse

0.14

èĢĲ

0.14

_kel

0.14

egra

0.13

Activations Density 0.048%

abbreviated references or initials followed by punctuation

No Comments

No Known Activations