INDEX

Explanations

terms related to raising doubts or questions

oai_token-act-pair · gpt-3.5-turbo

phrases associated with questioning and doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

omorphic

-0.71

oked

-0.71

cknow

-0.69

approved

-0.67

©¶æ

-0.66

married

-0.65

 completed

-0.64

 guided

-0.61

 fing

-0.61

 perfected

-0.61

POSITIVE LOGITS

 havoc

0.85

"]=>

0.76

Marginal

0.70

 Logged

0.69

 certs

0.67

 GOODMAN

0.66

 Skydragon

0.66

 Flavoring

0.64

 unnecessarily

0.64

çīĪ

0.64

Activations Density 0.274%

terms related to raising doubts or questions

phrases associated with questioning and doubt

No Comments

No Known Activations