INDEX

Explanations

various forms of the word "cum" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ibel

-0.17

izio

-0.15

ÏĩÎ®

-0.15

pod

-0.15

pine

-0.15

actory

-0.15

iembre

-0.14

pst

-0.14

amento

-0.14

POSITIVE LOGITS

ulative

0.30

mins

0.26

ulus

0.25

quat

0.24

ming

0.23

bersome

0.22

Cum

0.18

uli

0.17

Cum

0.17

ulating

0.17

Activations Density 0.006%

various forms of the word "cum" in different contexts

No Comments

No Known Activations

various forms of the word "cum" in different contexts

No Comments

No Known Activations