INDEX

Explanations

phrases related to giving advice and support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ANTE

-0.17

ante

-0.16

urch

-0.16

Ib

-0.14

éīĦ

-0.14

 vice

-0.14

Fri

-0.14

wal

-0.14

poz

-0.13

ken

-0.13

POSITIVE LOGITS

 similar

0.22

 similarly

0.21

similar

0.18

odyn

0.17

imilar

0.17

 benzer

0.16

ÃŃt

0.16

Ø§Ø¨Ø±

0.15

anders

0.15

.tie

0.15

Activations Density 0.079%

phrases related to giving advice and support

No Comments

No Known Activations