INDEX

Explanations

specific terms and concepts indicative of technical or scientific frameworks and categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

rary

-0.16

ramids

-0.16

_MPI

-0.15

åŃĺäºİ

-0.15

getQuery

-0.14

 Benchmark

-0.14

ADED

-0.14

emailer

-0.14

leen

-0.14

POSITIVE LOGITS

acon

0.15

ectar

0.15

anut

0.14

ä¹Ī

0.14

æ»ħ

0.14

afi

0.13

 *)((

0.13

 ãĢ

0.13

çĹ

0.13

ingo

0.13

Activations Density 0.005%

specific terms and concepts indicative of technical or scientific frameworks and categories

No Comments

No Known Activations