INDEX

Explanations

mentionings of words related to recognition or identifiability

oai_token-act-pair · gpt-3.5-turbo

words related to recognition and acknowledgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Grind

-0.74

 DRAG

-0.69

 setbacks

-0.67

 Tone

-0.65

 Tempest

-0.63

 sliding

-0.62

 Diver

-0.62

 abortions

-0.61

 delaying

-0.61

 reaction

-0.60

POSITIVE LOGITS

isable

1.36

izable

1.34

isance

1.23

recogn

1.20

ibly

1.06

ition

1.06

ances

1.05

ational

1.03

ising

1.01

usable

1.00

Activations Density 0.013%

mentionings of words related to recognition or identifiability

words related to recognition and acknowledgment

No Comments

No Known Activations