INDEX

Explanations

phrases that contain multiple uses of the word "in" within a context of analysis or description

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

abay

-0.16

_exchange

-0.15

 fact

-0.15

Exchange

-0.15

FACT

-0.15

cie

-0.14

ambi

-0.14

åĢ

-0.14

 spite

-0.13

fact

-0.13

POSITIVE LOGITS

 detail

0.61

detail

0.49

 depth

0.47

 Detail

0.43

 greater

0.42

-detail

0.40

Detail

0.39

 great

0.36

depth

0.36

 detalle

0.35

Activations Density 0.160%

phrases that contain multiple uses of the word "in" within a context of analysis or description

No Comments

No Known Activations