INDEX

Explanations

phrases or terms related to assessments of significance or impact in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

heimer

-0.17

 bunch

-0.15

acades

-0.15

 array

-0.14

isk

-0.14

iam

-0.13

 abundance

-0.13

 backing

-0.13

 Holland

-0.13

ahoma

-0.13

POSITIVE LOGITS

 factor

0.17

 source

0.16

key

0.15

source

0.15

factor

0.14

 prime

0.14

 valid

0.14

ÄĽÅ¾

0.14

 buzz

0.14

 form

0.14

Activations Density 0.143%

phrases or terms related to assessments of significance or impact in various contexts

No Comments

No Known Activations