INDEX

Explanations

summaries and their relevance in content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ð°Ð»ÐµÐ¶

-0.16

ad

-0.15

ieri

-0.14

ucher

-0.14

ç½ļ

-0.14

ê³³

-0.14

Ã¤ter

-0.14

uml

-0.13

enz

-0.13

Choices

-0.13

POSITIVE LOGITS

enance

0.17

OfWork

0.15

-ÑĤÐ¾

0.14

 Bene

0.14

oftware

0.14

ative

0.14

izar

0.14

Ð´Ð°Ð¼

0.14

ird

0.14

hin

0.14

Activations Density 0.030%

summaries and their relevance in content

No Comments

No Known Activations