INDEX

Explanations

expressions of appreciation

oai_token-act-pair · gpt-3.5-turbo

references to appreciation and acknowledgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 tactics

-0.73

 calculus

-0.73

GD

-0.71

 Primordial

-0.71

tone

-0.70

FactoryReloaded

-0.69

 defenses

-0.67

 strategy

-0.67

 Feld

-0.65

ford

-0.64

POSITIVE LOGITS

izable

1.07

umerable

1.05

atively

0.97

atable

0.93

ators

0.93

uers

0.93

urious

0.92

icably

0.92

ator

0.91

ision

0.91

Activations Density 0.019%

expressions of appreciation

references to appreciation and acknowledgment

No Comments

No Known Activations