INDEX

Explanations

references to transitions and changes in systems or states

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

ANCE

-0.15

.GetSize

-0.14

 impr

-0.14

 grav

-0.14

TD

-0.14

 projections

-0.13

 Knox

-0.13

 Hamm

-0.13

plet

-0.13

POSITIVE LOGITS

 transition

0.35

transition

0.33

 Transition

0.33

-transition

0.30

Smooth

0.29

Transition

0.29

 transitional

0.28

 abrupt

0.28

 transitions

0.28

 smooth

0.27

Activations Density 0.161%

references to transitions and changes in systems or states

No Comments

No Known Activations