INDEX

Explanations

expressions of uncertainty or questions about intentions and desires

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

alat

-0.17

abant

-0.16

nah

-0.16

heimer

-0.15

crest

-0.15

NAS

-0.15

_pan

-0.14

 otherwise

-0.14

 sÃ¥

-0.14

ë¥ĺ

-0.14

POSITIVE LOGITS

 Hint

0.18

 maybe

0.16

_blend

0.15

 Certainly

0.15

Hint

0.15

ighted

0.14

ãĤ¿ãĥ«

0.14

Bid

0.14

hint

0.14

 certainly

0.14

Activations Density 0.185%

expressions of uncertainty or questions about intentions and desires

No Comments

No Known Activations