INDEX

Explanations

questions and statements regarding the validity of claims and truths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kop

-0.15

 Authority

-0.14

 Unified

-0.14

.bio

-0.14

Ð¼Ñı

-0.14

ÐµÑĢÐµÐ¶

-0.13

van

-0.13

.attrs

-0.13

asm

-0.13

pedia

-0.13

POSITIVE LOGITS

 true

0.84

true

0.76

 TRUE

0.66

 True

0.65

True

0.61

TRUE

0.60

 truth

0.54

	true

0.54

(true

0.52

 verdade

0.52

Activations Density 0.317%

questions and statements regarding the validity of claims and truths

No Comments

No Known Activations