INDEX

Explanations

mentions of web-related terms and structures

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

regnum

-0.15

haven

-0.14

anut

-0.14

erule

-0.14

pong

-0.14

KF

-0.14

steen

-0.14

Å¡et

-0.14

 exercitation

-0.14

geist

-0.14

POSITIVE LOGITS

ugas

0.17

(DBG

0.15

ESA

0.14

 Stanton

0.14

roe

0.14

 numberWith

0.14

 clad

0.13

 ↵        ↵

0.13

.scalar

0.13

Activations Density 0.021%

mentions of web-related terms and structures

No Comments

No Known Activations