INDEX

Explanations

phrases that introduce reported speech or references to information sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

-0.16

ukan

-0.15

Ð¼ÐµÐ½

-0.15

iversity

-0.15

ortal

-0.15

urum

-0.15

leo

-0.14

psc

-0.14

raison

-0.14

yb

-0.14

POSITIVE LOGITS

ly

0.28

ately

0.20

 according

0.18

LY

0.18

ally

0.17

ÑģÑĮ

0.17

alf

0.17

ingly

0.16

æĸ¼

0.15

oun

0.15

Activations Density 0.042%

phrases that introduce reported speech or references to information sources

No Comments

No Known Activations