INDEX

Explanations

quotes and dialogue in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 disb

-0.16

ãĥ¼ãĥĲ

-0.15

ideographic

-0.15

ãĥ¼ãĤ¹ãĥĪ

-0.15

à¯įà®

-0.15

ugi

-0.14

PRS

-0.14

iÐ²

-0.14

nud

-0.14

ENA

-0.14

POSITIVE LOGITS

Zap

0.15

loat

0.15

aram

0.15

set

0.15

amble

0.15

Ã©rique

0.15

amilia

0.14

Ð¾Ð»ÑĮÐ½Ð¾

0.14

reek

0.14

ener

0.14

Activations Density 0.044%

quotes and dialogue in the text

No Comments

No Known Activations