INDEX

Explanations

instances where the action of tending to something or someone is described

oai_token-act-pair · gpt-3.5-turbo

expressions related to the concept of "tending" or caring for something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Kl

-0.76

hal

-0.67

Nak

-0.64

plan

-0.64

ania

-0.62

dash

-0.62

Tek

-0.62

fil

-0.62

Zel

-0.62

pat

-0.62

POSITIVE LOGITS

entimes

1.10

rils

0.95

pmwiki

0.94

 tremend

0.92

 tended

0.91

 assum

0.90

 conduc

0.87

 suscept

0.86

 occas

0.86

 encount

0.85

Activations Density 0.007%

instances where the action of tending to something or someone is described

expressions related to the concept of "tending" or caring for something

No Comments

No Known Activations