INDEX

Explanations

dates in specific formats

oai_token-act-pair · gpt-3.5-turbo

prepositions indicating time and location

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 clipboard

-0.72

ank

-0.67

 abound

-0.63

 simpl

-0.61

 categor

-0.59

 cheated

-0.59

advertisement

-0.59

 originated

-0.58

icator

-0.58

 derivative

-0.58

POSITIVE LOGITS

igslist

0.73

 Nanto

0.73

 Maiden

0.73

rocal

0.72

ISON

0.71

 Tokyo

0.66

 Steel

0.62

 JUSTICE

0.62

enger

0.62

roc

0.61

Activations Density 0.295%

dates in specific formats

prepositions indicating time and location

No Comments

No Known Activations