INDEX

Explanations

personal pronouns followed by verbs indicating direction or movement

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and their locations or intentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĸļ

-0.65

uthor

-0.65

Reviewer

-0.65

advertisement

-0.64

naires

-0.64

odor

-0.61

 Leilan

-0.61

chan

-0.61

oret

-0.60

ware

-0.60

POSITIVE LOGITS

 originated

0.93

 reside

0.90

 belong

0.89

 originate

0.88

 resides

0.85

 resided

0.84

 geographically

0.82

 origin

0.80

 weakest

0.80

 belongs

0.79

Activations Density 0.130%

personal pronouns followed by verbs indicating direction or movement

references to specific individuals and their locations or intentions

No Comments

No Known Activations