INDEX

Explanations

phrases related to the causes or factors leading to certain outcomes

oai_token-act-pair · gpt-3.5-turbo

Terms related to significant causes or contributing factors 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 @ 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

atomic

-0.72

 conventions

-0.65

 Tours

-0.64

 orbits

-0.64

assies

-0.62

Å¡

-0.61

 tours

-0.61

arij

-0.58

ancies

-0.58

 spiders

-0.57

POSITIVE LOGITS

 Catalyst

0.81

 deterrent

0.79

 GOODMAN

0.76

 contributor

0.75

ICLE

0.69

acea

0.69

SourceFile

0.68

 motiv

0.67

 motivating

0.66

ainer

0.66

Activations Density 0.284%

phrases related to the causes or factors leading to certain outcomes

Terms related to significant causes or contributing factors in various contexts

No Comments

No Known Activations