INDEX

Explanations

exceptions to rules or norms

oai_token-act-pair · gpt-3.5-turbo

references to exceptions or cases that diverge from a rule

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

igion

-0.71

uliffe

-0.70

ongyang

-0.70

 sqor

-0.69

zeb

-0.65

 degrading

-0.64

 IMAGES

-0.63

 residues

-0.62

FORMATION

-0.62

 sacrific

-0.62

POSITIVE LOGITS

 exceptions

0.80

 exception

0.75

 breed

0.71

 rule

0.71

 altogether

0.70

Exception

0.66

 occasional

0.65

arily

0.65

anke

0.65

ensis

0.64

Activations Density 0.155%

exceptions to rules or norms

references to exceptions or cases that diverge from a rule

No Comments

No Known Activations