INDEX

Explanations

first instances or beginnings of events or entities

oai_token-act-pair · gpt-3.5-turbo

phrases referring to first rounds or initial stages in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

beans

-0.73

 Twins

-0.65

gans

-0.64

seys

-0.63

 masc

-0.61

clips

-0.61

 Animals

-0.60

 launchers

-0.60

asus

-0.59

 Characters

-0.59

POSITIVE LOGITS

 approximation

0.89

 glim

0.79

0.77

 glimpse

0.77

 contender

0.73

 impression

0.73

class

0.72

 dimensional

0.72

Rate

0.71

stroke

0.70

Activations Density 0.115%

first instances or beginnings of events or entities

phrases referring to first rounds or initial stages in various contexts

No Comments

No Known Activations