INDEX

Explanations

references to trailers and video content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 sÃ¡t

-0.16

ÐµÑĢÐ±

-0.14

 Illustrated

-0.14

 Glob

-0.13

 nesta

-0.13

aku

-0.13

zzo

-0.12

aea

-0.12

visor

-0.12

.aspx

-0.12

POSITIVE LOGITS

Embedded

0.25

 below

0.23

 embedded

0.23

 embed

0.22

below

0.21

 Embedded

0.21

embedded

0.21

 AFTER

0.21

Embed

0.20

 Embed

0.19

Activations Density 0.074%

references to trailers and video content

No Comments

No Known Activations