INDEX

Explanations

phrases indicating surprise or unexpected realizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

alo

-0.14

lue

-0.13

nock

-0.13

lej

-0.13

eated

-0.13

ÑĮÑİ

-0.13

getti

-0.12

nul

-0.12

/single

-0.12

orang

-0.12

POSITIVE LOGITS

 thought

0.48

 expected

0.43

thought

0.42

 Thought

0.42

expected

0.37

Thought

0.36

 anticipated

0.34

 hoped

0.33

Expected

0.33

 assumed

0.33

Activations Density 0.362%

phrases indicating surprise or unexpected realizations

No Comments

No Known Activations