INDEX

Explanations

keywords and properties related to programming structures and configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 muht

-0.15

 sucker

-0.15

Bam

-0.14

mostat

-0.14

UTTON

-0.14

KN

-0.14

 Grat

-0.14

 Datum

-0.13

ouston

-0.13

 unthinkable

-0.13

POSITIVE LOGITS

ones

0.17

Ð½Ð¸ÑĨÐ°

0.16

_exempt

0.16

 ones

0.14

_isr

0.14

riel

0.14

å¤®

0.14

ullo

0.14

idl

0.14

enza

0.14

Activations Density 0.118%

keywords and properties related to programming structures and configurations

No Comments

No Known Activations