INDEX

Explanations

phrases that indicate drawing inspiration or connections from various sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ovah

-0.18

ature

-0.17

udas

-0.16

iment

-0.15

aya

-0.15

oy

-0.15

amas

-0.14

oyal

-0.14

è½½

-0.14

iments

-0.14

POSITIVE LOGITS

 draw

0.31

 draws

0.30

 drawn

0.30

 attention

0.30

 Draw

0.28

 Draws

0.28

.Draw

0.27

attention

0.27

.draw

0.26

 drew

0.26

Activations Density 0.021%

phrases that indicate drawing inspiration or connections from various sources

No Comments

No Known Activations