INDEX

Explanations

phrases that indicate possession or experience, particularly with the verb "has"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

-0.16

å¥ı

-0.15

eve

-0.15

sm

-0.15

uze

-0.14

PERT

-0.14

Ø³ØªØ§ÙĨÛĮ

-0.14

 Temmuz

-0.14

ILING

-0.13

ÃŁen

-0.13

POSITIVE LOGITS

ĵ¨

0.15

 CONTROL

0.14

avana

0.14

otron

0.14

iable

0.14

ayscale

0.13

erli

0.13

kov

0.13

itals

0.13

otto

0.13

Activations Density 0.039%

phrases that indicate possession or experience, particularly with the verb "has"

No Comments

No Known Activations