INDEX

Explanations

elements and structures in code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

bum

-0.15

-urlencoded

-0.15

æŁĦ

-0.15

à¤Ĥà¤¡à¤²

-0.14

à¸¢à¸¡

-0.14

bers

-0.13

 Leban

-0.13

DMA

-0.13

alone

-0.13

avec

-0.13

POSITIVE LOGITS

ëľ

0.17

atsu

0.16

ettes

0.15

ships

0.15

neys

0.15

rike

0.14

IRST

0.14

osex

0.14

typings

0.14

inema

0.14

Activations Density 0.262%

elements and structures in code snippets

No Comments

No Known Activations