INDEX

Explanations

terms associated with legends and legendary status

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

erman

-0.16

illon

-0.16

angent

-0.15

age

-0.15

êµ¬

-0.15

istes

-0.15

gv

-0.15

ordes

-0.14

NSNotification

-0.14

chop

-0.14

POSITIVE LOGITS

ede

0.19

enes

0.15

 sitesi

0.15

resher

0.15

UNET

0.14

oard

0.14

utzer

0.14

iterr

0.14

ContentType

0.13

oui

0.13

Activations Density 0.019%

terms associated with legends and legendary status

No Comments

No Known Activations