INDEX

Explanations

references to recent works or updates

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

ered

-0.16

æģ

-0.15

closure

-0.15

eric

-0.15

 closure

-0.15

Ã£n

-0.15

éĸĢ

-0.14

çĶ³

-0.14

nar

-0.14

POSITIVE LOGITS

/latest

0.19

Ð¾ÑĢÐ°Ð»ÑĮ

0.17

icks

0.15

ê·¹

0.14

ÅĽcie

0.14

ace

0.13

isode

0.13

 diseÃ±ador

0.13

iley

0.13

ests

0.13

Activations Density 0.012%

references to recent works or updates

No Comments

No Known Activations