INDEX

Explanations

sections of text with headings or topics that indicate a categorization or classification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

linger

-0.19

esch

-0.14

{{{

-0.14

.commons

-0.13

imps

-0.13

berries

-0.13

 Mash

-0.13

ÑĢÐ¾ÑģÑĤÐ¾

-0.13

Je

-0.13

berry

-0.13

POSITIVE LOGITS

illet

0.17

*(*

0.16

eyse

0.14

zos

0.14

essel

0.14

aket

0.14

 intermedi

0.14

Ð¼Ð¸ÑĢ

0.13

amak

0.13

 Ticaret

0.13

Activations Density 0.020%

sections of text with headings or topics that indicate a categorization or classification

No Comments

No Known Activations