INDEX

Explanations

instances of the word "invocations"

oai_token-act-pair · gpt-3.5-turbo

terms related to invasion and intrusion, particularly in a legal or societal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

hyde

-0.87

WARE

-0.75

Ĥ¬

-0.73

FINEST

-0.67

 hydrogen

-0.65

 fixme

-0.65

geon

-0.64

GGGG

-0.64

 Judicial

-0.62

ovych

-0.62

POSITIVE LOGITS

igor

1.23

itational

1.21

itations

1.14

asion

1.08

ocations

1.07

okes

1.06

asions

1.05

idious

1.02

ulnerable

1.00

itures

1.00

Activations Density 0.009%

instances of the word "invocations"

terms related to invasion and intrusion, particularly in a legal or societal context

No Comments

No Known Activations