INDEX

Explanations

references to legal timelines or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

_atual

-0.15

Ãºng

-0.15

rios

-0.14

anse

-0.14

alue

-0.14

ivr

-0.14

ptype

-0.14

ramework

-0.14

 Collapse

-0.14

 halftime

-0.14

POSITIVE LOGITS

ë¶ĢíĦ°

0.20

 earliest

0.18

croft

0.16

annon

0.16

ãĥ¼ãĥĦ

0.15

aja

0.15

Tod

0.14

ider

0.14

arian

0.14

Activations Density 0.225%

references to legal timelines or processes

No Comments

No Known Activations