INDEX

Explanations

action verbs and terms related to progress or achievement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

HAV

-0.15

 Monte

-0.14

folio

-0.14

okus

-0.14

.scalablytyped

-0.14

sur

-0.14

 delicate

-0.13

Montserrat

-0.13

 Roberts

-0.13

POSITIVE LOGITS

usercontent

0.17

izr

0.17

jit

0.17

entication

0.16

osite

0.15

raison

0.15

oring

0.15

istrat

0.15

Ã³st

0.14

ubit

0.14

Activations Density 0.184%

action verbs and terms related to progress or achievement

No Comments

No Known Activations