INDEX

Explanations

references to concepts and elements related to behind-the-scenes narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

uman

-0.17

ÐµÑĢÐ¿

-0.15

affer

-0.14

lew

-0.14

adius

-0.14

ERA

-0.13

uite

-0.13

 crude

-0.13

æĶ

-0.13

 verg

-0.13

POSITIVE LOGITS

çħ§

0.16

burn

0.16

MMdd

0.15

-Allow

0.15

ÑĢÑĥÐ¶

0.15

kers

0.15

rosso

0.15

ãĥ¬ãĥ¼

0.14

Charsets

0.14

/Dk

0.14

Activations Density 0.018%

references to concepts and elements related to behind-the-scenes narratives

No Comments

No Known Activations