INDEX

Explanations

categories and terms related to various forms of media and medical conditions

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

ses

-0.14

co

-0.14

arat

-0.14

ucas

-0.14

ates

-0.14

iez

-0.13

 Ding

-0.13

ÑĥÐ½Ð´

-0.13

charm

-0.13

\@

-0.13

POSITIVE LOGITS

nad

0.14

okino

0.14

chter

0.13

weeted

0.13

lant

0.13

INLINE

0.13

ibName

0.13

asmus

0.13

ynos

0.13

olean

0.13

Activations Density 0.086%

categories and terms related to various forms of media and medical conditions

No Comments

No Known Activations