INDEX

Explanations

references to specific locations, particularly cities and academic publishers

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

ombs

-0.14

 mixes

-0.14

alla

-0.14

oms

-0.14

ebe

-0.14

zens

-0.13

 Practical

-0.13

zew

-0.13

ter

-0.13

pon

-0.13

POSITIVE LOGITS

ATAB

0.17

chwitz

0.16

istrovstvÃŃ

0.16

ãĥªãĥ¼ãĤº

0.15

klady

0.15

istros

0.15

FileVersion

0.15

Interop

0.15

iginal

0.14

 EntityState

0.14

Activations Density 0.183%

references to specific locations, particularly cities and academic publishers

No Comments

No Known Activations