INDEX

Explanations

phrases related to the presence and evaluation of evidence in arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ë²Ī

-0.15

andro

-0.15

lek

-0.14

/umd

-0.13

adol

-0.13

ottle

-0.13

IPA

-0.13

lesai

-0.13

alert

-0.13

POSITIVE LOGITS

 evidence

0.84

 Evidence

0.69

Evidence

0.65

 proof

0.52

vidence

0.51

 evid

0.50

 Proof

0.39

è¯ģ

0.39

proof

0.38

Proof

0.37

Activations Density 0.305%

phrases related to the presence and evaluation of evidence in arguments

No Comments

No Known Activations