INDEX

Explanations

negations or phrases indicating something not occurring

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-sidebar

-0.17

pur

-0.15

pson

-0.14

ektor

-0.14

tsx

-0.14

subj

-0.14

Yap

-0.14

Ð½Ñİ

-0.14

 amounts

-0.13

 Diss

-0.13

POSITIVE LOGITS

wei

0.15

aylor

0.14

bine

0.14

 Ø§ÙĦÙħÙĪØ³

0.14

ActionCreators

0.14

-normal

0.14

è©±

0.14

elite

0.13

quir

0.13

Activations Density 0.013%

negations or phrases indicating something not occurring

No Comments

No Known Activations