INDEX

Explanations

assertions or claims related to legal or contentious issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

eya

-0.15

 Mellon

-0.15

usan

-0.15

ester

-0.14

fel

-0.14

favor

-0.14

 ÑĥÐ¶

-0.14

Î³ÎºÎ±

-0.14

ovy

-0.14

uyá»ĩn

-0.14

POSITIVE LOGITS

 rational

0.16

errick

0.15

rica

0.15

ĵn

0.15

 that

0.14

ç¼

0.14

Cheers

0.14

urb

0.14

rvine

0.13

à¥Įà¤Ĥ

0.13

Activations Density 0.283%

assertions or claims related to legal or contentious issues

No Comments

No Known Activations