INDEX

Explanations

television show titles and associated ratings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

à¸¥à¸²

-0.16

ff

-0.15

hl

-0.15

ymi

-0.14

 Latter

-0.14

ae

-0.14

hill

-0.13

nex

-0.13

imer

-0.13

POSITIVE LOGITS

Ð·Ñĭ

0.15

agit

0.15

amam

0.15

åĺĽ

0.15

anca

0.15

ella

0.14

okane

0.14

cheid

0.14

arium

0.14

boo

0.14

Activations Density 0.038%

television show titles and associated ratings

No Comments

No Known Activations