INDEX

Explanations

instances of the contraction "I'd."

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

Negative Logits

gles

-0.16

é¼»

-0.15

aupt

-0.14

phinx

-0.14

 moons

-0.14

plets

-0.14

aits

-0.14

tring

-0.14

fant

-0.13

REA

-0.13

POSITIVE LOGITS

Mob

0.16

Mob

0.15

our

0.14

oter

0.14

iom

0.14

<*

0.14

hasil

0.14

UNE

0.14

ibil

0.14

 politic

0.14

Activations Density 0.016%

instances of the contraction "I'd."

No Known Activations