INDEX

Explanations

instances of names or significant identifiers in a context, particularly those that indicate prominence or importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ãĥ£

-0.17

istik

-0.16

Ø§Ø²Ùĩ

-0.16

 allegedly

-0.15

another

-0.14

xs

-0.14

ones

-0.14

ie

-0.13

arda

-0.13

OOK

-0.13

POSITIVE LOGITS

 which

0.32

which

0.26

 whose

0.26

 Which

0.25

Which

0.23

.which

0.21

cui

0.21

 WHICH

0.21

whose

0.20

who

0.19

Activations Density 0.042%

instances of names or significant identifiers in a context, particularly those that indicate prominence or importance

No Comments

No Known Activations