INDEX

Explanations

terms related to associations and connections, particularly in the context of relationships and influences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

anko

-0.17

éłŃ

-0.16

nf

-0.15

odi

-0.15

vv

-0.15

ograms

-0.15

interaction

-0.15

 Sher

-0.15

ancia

-0.14

inters

-0.14

POSITIVE LOGITS

 closely

0.19

 directly

0.19

 with

0.19

 associate

0.19

 associated

0.18

 assoc

0.18

 linked

0.18

 associ

0.17

 tied

0.17

associate

0.17

Activations Density 0.047%

terms related to associations and connections, particularly in the context of relationships and influences

No Comments

No Known Activations