INDEX

Explanations

repeated numerical references or scores related to sporting events

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

ackle

-0.17

ieme

-0.15

airs

-0.15

ffa

-0.14

iske

-0.14

 conduct

-0.14

oje

-0.14

vious

-0.14

kad

-0.14

pson

-0.13

POSITIVE LOGITS

illez

0.16

á»ĩu

0.15

ialis

0.14

å®³

0.14

ually

0.14

ertype

0.14

/GPL

0.14

rowave

0.14

erie

0.14

/plain

0.14

Activations Density 0.004%

repeated numerical references or scores related to sporting events

No Comments

No Known Activations