INDEX

Explanations

proper nouns related to a particular place or location

oai_token-act-pair · gpt-3.5-turbo

references to specific geographical locations or entities, particularly related to Sonoma

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

balance

-1.06

anooga

-0.84

reb

-0.70

enser

-0.69

rador

-0.69

ilian

-0.67

 intuitive

-0.67

drivers

-0.67

book

-0.67

wick

-0.67

POSITIVE LOGITS

oma

1.02

Tos

0.79

pha

0.77

qua

0.74

 Cancer

0.74

 Luxem

0.73

 Nost

0.72

 Mayhem

0.72

 Linda

0.71

Naz

0.71

Activations Density 0.015%

proper nouns related to a particular place or location

references to specific geographical locations or entities, particularly related to Sonoma

No Comments

No Known Activations