INDEX

Explanations

questions being asked or referred to in a text

oai_token-act-pair · gpt-3.5-turbo

instances of "questions" and their context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ufact

-0.96

rites

-0.74

ortunately

-0.73

ensions

-0.68

gradation

-0.68

orpor

-0.68

©¶æ¥µ

-0.66

âĶģ

-0.66

 Tycoon

-0.66

oreal

-0.65

POSITIVE LOGITS

naires

1.43

naire

1.31

 answered

1.23

 posed

1.07

 answ

0.99

 asked

0.98

answered

0.98

answer

0.95

 questions

0.92

Answer

0.92

Activations Density 0.043%

questions being asked or referred to in a text

instances of "questions" and their context

No Comments

No Known Activations