INDEX

Explanations

instances where "in" is used to indicate a location or context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Ellen

-0.15

iks

-0.14

tes

-0.14

ebek

-0.14

 Burke

-0.14

baum

-0.14

ettes

-0.14

ErrorHandler

-0.14

 Stark

-0.14

 Rain

-0.13

POSITIVE LOGITS

æķ·

0.16

 ==============================================================

0.14

chema

0.14

gc

0.14

Å¾ila

0.14

/GPL

0.14

 IOCTL

0.14

andel

0.14

opia

0.14

 awakeFromNib

0.13

Activations Density 0.112%

instances where "in" is used to indicate a location or context

No Comments

No Known Activations