INDEX

Explanations

expressions and terms related to intentional actions and deliberate choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

jem

-0.14

assi

-0.14

lder

-0.14

 Ø³Ú©

-0.14

Ð¾Ð²Ñĥ

-0.14

ipar

-0.14

_TC

-0.13

 blessing

-0.13

å±±å¸Ĥ

-0.13

.AspNet

-0.13

POSITIVE LOGITS

uja

0.16

culate

0.16

 Dome

0.15

idd

0.14

icros

0.14

kaar

0.13

ibase

0.13

urf

0.13

ode

0.13

Activations Density 0.001%

expressions and terms related to intentional actions and deliberate choices

No Comments

No Known Activations