INDEX

Explanations

technical terms related to software and technology

oai_token-act-pair · gpt-3.5-turbo

terms related to various forms of data management and efficiency in systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Ø§ÙĦ

-0.81

AME

-0.71

Recomm

-0.69

Ire

-0.69

Untitled

-0.68

$.

-0.68

ayn

-0.67

brance

-0.67

ufact

-0.67

incible

-0.66

POSITIVE LOGITS

 aside

0.75

huh

0.72

 refers

0.70

 meanwhile

0.67

 spoilers

0.66

 consists

0.64

 ---------

0.64

 includes

0.64

 Sonia

0.63

 --------

0.63

Activations Density 0.560%

technical terms related to software and technology

terms related to various forms of data management and efficiency in systems

No Comments

No Known Activations