INDEX

Explanations

statements of evaluation or judgment regarding value

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

illac

-0.16

rud

-0.16

psc

-0.15

ichick

-0.15

peare

-0.14

weise

-0.14

OnError

-0.14

lude

-0.14

ahren

-0.13

omatic

-0.13

POSITIVE LOGITS

Äįen

0.15

 Whereas

0.15

Wen

0.15

WN

0.14

 Ð¶Ðµ

0.14

ucken

0.14

INCLUDED

0.13

çŀ

0.13

á»įc

0.13

ÑĢÐ¾Ñĩ

0.13

Activations Density 0.000%

statements of evaluation or judgment regarding value

No Comments

No Known Activations