INDEX

Explanations

mentions of a specific name "Till" with varying levels of relevance

oai_token-act-pair · gpt-3.5-turbo

references to the word "Till" and its variations 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 @ 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

VERTISEMENT

-0.87

TABLE

-0.78

shapeshifter

-0.70

 SetTextColor

-0.69

VID

-0.68

MODE

-0.66

BOOK

-0.64

pron

-0.62

journal

-0.61

Premium

-0.60

POSITIVE LOGITS

 Till

1.20

Til

0.97

 Nadu

0.90

terson

0.77

kees

0.76

sten

0.76

iband

0.76

osaurus

0.74

burgh

0.74

mann

0.73

Activations Density 0.004%

mentions of a specific name "Till" with varying levels of relevance

references to the word "Till" and its variations in various contexts

No Comments

No Known Activations