INDEX

Explanations

phrases and expressions that signal exceptions or qualifications in statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ãģĦãĤĦ

-0.18

-0.16

HEME

-0.16

ntag

-0.15

 Projection

-0.14

ibri

-0.14

onen

-0.14

onia

-0.14

Pure

-0.14

POSITIVE LOGITS

ehr

0.16

/not

0.15

icl

0.15

.builders

0.15

DefaultValue

0.14

 anymore

0.14

icus

0.14

ecs

0.14

Cul

0.14

izable

0.13

Activations Density 0.082%

phrases and expressions that signal exceptions or qualifications in statements

No Comments

No Known Activations