INDEX

Explanations

references to obstructive behaviors and situations

oai_token-act-pair · gpt-3.5-turbo

terms related to obstruction or hindrance, particularly in a legal or bureaucratic context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ndra

-0.72

 Grind

-0.71

CHR

-0.69

mut

-0.69

 Coat

-0.69

FI

-0.68

 Soph

-0.68

 Stras

-0.68

www

-0.67

tan

-0.67

POSITIVE LOGITS

 obstruction

1.32

 obstruct

1.14

structed

0.93

struction

0.87

 undermin

0.84

 sabot

0.82

avorite

0.81

agall

0.80

uously

0.78

 overshadow

0.78

Activations Density 0.013%

references to obstructive behaviors and situations

terms related to obstruction or hindrance, particularly in a legal or bureaucratic context

No Comments

No Known Activations