INDEX

Explanations

words related to hints or signs suggesting a particular course of action or understanding

oai_token-act-pair · gpt-3.5-turbo

phrases that suggest evidence or signs of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

iless

-0.74

ran

-0.72

}}

-0.66

rio

-0.63

finals

-0.62

js

-0.62

jo

-0.61

ventures

-0.61

Won

-0.58

 Prosper

-0.58

POSITIVE LOGITS

 indication

3.53

 indications

2.67

 indicator

2.23

 indicators

1.57

 indicative

1.43

 clue

1.34

 indicating

1.33

 hint

1.31

 acknowledgement

1.29

 acknowledgment

1.28

Activations Density 0.017%

words related to hints or signs suggesting a particular course of action or understanding

phrases that suggest evidence or signs of something

No Comments

No Known Activations