INDEX

Explanations

instances of specific phrases or terms related to evidence and claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.Undef

-0.16

styleType

-0.15

EMPLARY

-0.15

paque

-0.15

alace

-0.14

ascar

-0.14

rias

-0.14

iefs

-0.14

Î³Î¬

-0.14

 Äĳá»¡

-0.14

POSITIVE LOGITS

 idea

0.44

 notion

0.39

 suggestion

0.34

 belief

0.32

 impression

0.32

idea

0.32

 prospect

0.31

 question

0.31

 possibility

0.30

 proposition

0.29

Activations Density 0.251%

instances of specific phrases or terms related to evidence and claims

No Comments

No Known Activations