INDEX

Explanations

negative constructions and expressions of inability or avoidance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 vÃ´

-0.16

 licensors

-0.15

ë¡ľëĵľ

-0.14

reet

-0.14

 indirectly

-0.14

adh

-0.14

 specially

-0.13

_closure

-0.13

Ð¸ÑģÐ»Ð¾Ð²

-0.13

arie

-0.13

POSITIVE LOGITS

 miss

0.30

 disappoint

0.25

 soon

0.25

miss

0.25

 misses

0.22

 fail

0.22

 disappointed

0.22

 missed

0.21

 tire

0.21

 Soon

0.21

Activations Density 0.135%

negative constructions and expressions of inability or avoidance

No Comments

No Known Activations