INDEX

Explanations

words related to limitations, restrictions, or boundaries

oai_token-act-pair · gpt-3.5-turbo

references to limitations or restrictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

estone

-0.73

 Honour

-0.71

uni

-0.68

joy

-0.67

story

-0.67

mberg

-0.67

tein

-0.65

hod

-0.63

************

-0.63

psc

-0.62

POSITIVE LOGITS

 constraints

1.24

 constraint

1.11

 constrained

1.01

 imposed

0.92

 restraints

0.90

 dictates

0.88

 pressures

0.86

 cooker

0.84

 besie

0.81

 restricts

0.76

Activations Density 0.010%

words related to limitations, restrictions, or boundaries

references to limitations or restrictions

No Comments

No Known Activations