INDEX

Explanations

mentions of scripts and script-related terminology

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

iser

-0.17

unta

-0.15

ziej

-0.15

orta

-0.15

erta

-0.15

azi

-0.14

rees

-0.14

 sole

-0.14

StateToProps

-0.14

POSITIVE LOGITS

Susp

0.17

oppable

0.17

ä¸ļ

0.15

-economic

0.14

lopedia

0.14

.ly

0.14

 offen

0.14

dio

0.13

ancode

0.13

ergic

0.13

Activations Density 0.017%

mentions of scripts and script-related terminology

No Comments

No Known Activations