INDEX

Explanations

instances of numerical values and their related contextual phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ãģĭ

-0.15

ires

-0.14

pine

-0.14

etc

-0.14

 meanwhile

-0.13

ä¸ģ

-0.13

ãĥ£

-0.13

 exper

-0.13

 counterfeit

-0.13

POSITIVE LOGITS

 which

0.25

which

0.21

uh

0.17

.which

0.17

 WHICH

0.17

cui

0.17

Which

0.16

 Which

0.16

piler

0.14

atoire

0.14

Activations Density 0.078%

instances of numerical values and their related contextual phrases

No Comments

No Known Activations