INDEX

Explanations

instances of the word "to."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ereotype

-0.15

enty

-0.14

ruz

-0.14

eyse

-0.14

/spec

-0.14

plode

-0.14

èĶµ

-0.14

rez

-0.13

_Callback

-0.13

udic

-0.13

POSITIVE LOGITS

0.15

Rav

0.15

 Levine

0.15

rit

0.14

anagan

0.14

utor

0.14

.Generated

0.14

iq

0.14

_interfaces

0.13

in

0.13

Activations Density 0.057%

instances of the word "to."

No Comments

No Known Activations