INDEX

Explanations

instances of the word "draw" and its variations, indicating an emphasis on depiction or representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

aison

-0.16

acious

-0.16

iments

-0.16

amas

-0.15

ãĥĥãĤ¯ãĤ¹

-0.15

ovah

-0.14

 Castillo

-0.14

iment

-0.14

hv

-0.14

ured

-0.14

POSITIVE LOGITS

 draw

0.25

 Draw

0.24

(draw

0.21

.draw

0.21

 drawn

0.21

.Draw

0.21

 Draws

0.21

 conclusion

0.20

 drawing

0.20

 DRAW

0.20

Activations Density 0.031%

instances of the word "draw" and its variations, indicating an emphasis on depiction or representation

No Comments

No Known Activations