INDEX

Explanations

conversational elements and expressions of personal experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

templ

-0.16

 Sand

-0.15

ehr

-0.15

villa

-0.15

 Bret

-0.14

Kun

-0.14

 Prote

-0.14

jom

-0.14

eniable

-0.14

Nie

-0.14

POSITIVE LOGITS

 because

0.17

because

0.17

ä¸ºäºĨ

0.16

 nháº±m

0.16

jsc

0.15

nict

0.15

Because

0.15

çĽ®ãĤĴ

0.15

inho

0.15

aby

0.14

Activations Density 0.235%

conversational elements and expressions of personal experience

No Comments

No Known Activations