INDEX

Explanations

phrases indicating actions involving visibility or access to information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

abra

-0.18

usalem

-0.16

elin

-0.15

zM

-0.15

-fontawesome

-0.15

.gf

-0.14

 euch

-0.14

ona

-0.14

ds

-0.14

imedia

-0.14

POSITIVE LOGITS

 Ð²ÑĸÑĤ

0.15

 Prescott

0.15

ypad

0.15

ombs

0.15

æ²ĥ

0.14

nested

0.14

éĢł

0.14

ixel

0.14

ynos

0.14

 McInt

0.14

Activations Density 0.014%

phrases indicating actions involving visibility or access to information

No Comments

No Known Activations