INDEX

Explanations

specific mentions of research studies

oai_token-act-pair · gpt-3.5-turbo

references to academic studies

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

 offic

-0.63

 blink

-0.60

 bonded

-0.60

 nostalgic

-0.59

 resonance

-0.59

 volatile

-0.59

 overloaded

-0.59

 inert

-0.59

 fitt

-0.58

 slightest

-0.58

POSITIVE LOGITS

study

0.88

udo

0.85

Study

0.84

iveness

0.77

ãĥĻ

0.73

UF

0.73

 methodology

0.72

 conducted

0.72

 abroad

0.70

ador

0.67

Activations Density 0.031%

specific mentions of research studies

references to academic studies

No Comments

No Known Activations