INDEX

Explanations

verbs indicating past actions or states

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

izens

-0.16

ÃŃrk

-0.15

rror

-0.15

 Created

-0.15

bjerg

-0.15

ushi

-0.14

eniable

-0.14

 Asked

-0.14

 erect

-0.14

bsolute

-0.14

POSITIVE LOGITS

met

0.31

 greeted

0.28

met

0.21

mar

0.21

wid

0.20

 resisted

0.20

Met

0.19

 accompanied

0.19

 approved

0.18

 widely

0.18

Activations Density 0.206%

verbs indicating past actions or states

No Comments

No Known Activations