INDEX

Explanations

words related to computing and signal processing

oai_token-act-pair · gpt-3.5-turbo

terms associated with specific measurements or quantifications in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Accountability

-0.68

 Izan

-0.64

 Parish

-0.64

 Subtle

-0.64

 Lines

-0.63

 Honour

-0.63

 Integrity

-0.59

Gaw

-0.59

QC

-0.59

 Guid

-0.59

POSITIVE LOGITS

ole

0.73

omorph

0.73

assing

0.72

Äĵ

0.70

iform

0.70

ect

0.69

lic

0.68

agra

0.67

asing

0.67

azine

0.67

Activations Density 0.373%

words related to computing and signal processing

terms associated with specific measurements or quantifications in various contexts

No Comments

No Known Activations