INDEX

Explanations

intensifiers or modifiers, particularly the word "so" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

incy

-0.15

apolis

-0.15

uncan

-0.15

 somewhat

-0.15

uae

-0.14

more

-0.14

pane

-0.14

rup

-0.14

gram

-0.14

phan

-0.14

POSITIVE LOGITS

oo

0.45

ooo

0.40

oooo

0.39

-called

0.38

 much

0.37

oooooooo

0.33

apy

0.29

oth

0.29

 Much

0.28

iled

0.28

Activations Density 0.046%

intensifiers or modifiers, particularly the word "so" and its variations

No Comments

No Known Activations