INDEX

Explanations

terms related to magic and fantasy elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

strup

-0.15

 Hole

-0.15

unte

-0.15

kest

-0.14

MLE

-0.14

thon

-0.14

bildung

-0.14

ro

-0.14

metic

-0.14

hole

-0.14

POSITIVE LOGITS

 McCart

0.15

 RequestMethod

0.15

anus

0.14

byt

0.14

uite

0.14

andr

0.14

orget

0.14

.Validate

0.13

_challenge

0.13

ovich

0.13

Activations Density 0.047%

terms related to magic and fantasy elements

No Comments

No Known Activations