INDEX

Explanations

occurrences of temporal references and actions in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 bergen

-0.17

olini

-0.17

 Dodd

-0.15

ioni

-0.14

tring

-0.14

ALS

-0.14

sthrough

-0.14

__$

-0.14

ican

-0.14

URI

-0.14

POSITIVE LOGITS

unt

0.49

 unit

0.45

unt

0.42

unit

0.38

-unit

0.36

’

0.35

Unt

0.34

 util

0.34

 Unit

0.34

Âłt

0.33

Activations Density 0.088%

occurrences of temporal references and actions in a narrative context

No Comments

No Known Activations