INDEX

Explanations

characters that are complex and exhibit a mix of strengths and flaws within a narrative

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

ould

-0.17

pars

-0.15

.FLAG

-0.15

æİ¥çĿĢ

-0.15

ÂŃi

-0.14

Ã¼f

-0.14

anked

-0.14

igger

-0.14

 meanwhile

-0.14

validators

-0.14

POSITIVE LOGITS

orida

0.16

icable

0.15

ÐºÐ¾Ð³

0.15

imoto

0.15

 ocas

0.15

egas

0.14

 deep

0.14

 notable

0.14

IME

0.14

 often

0.14

Activations Density 0.017%

characters that are complex and exhibit a mix of strengths and flaws within a narrative

No Comments

No Known Activations