INDEX

Explanations

punctuation and numerical values in the document

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

ignon

-0.15

èİ

-0.15

ectar

-0.15

.iv

-0.14

 corrid

-0.14

ront

-0.13

acket

-0.13

oÅ¾

-0.13

ãĤ¿ãĥ³

-0.13

 jadx

-0.13

POSITIVE LOGITS

ANCELED

0.14

¶ļ

0.14

Sit

0.13

 Bash

0.13

 others

0.13

æħİ

0.13

 Mine

0.13

 hers

0.12

 spins

0.12

Activations Density 0.050%

punctuation and numerical values in the document

No Comments

No Known Activations