INDEX

Explanations

mentions of geographical locations with the term "Cape" in them

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Cape" in relation to various places

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

arial

-0.82

iances

-0.76

arily

-0.74

amine

-0.73

omen

-0.72

orno

-0.70

naires

-0.69

agi

-0.68

atically

-0.68

mble

-0.68

POSITIVE LOGITS

llan

1.12

lla

0.99

Cod

0.97

 Canaveral

0.97

 Coral

0.85

 Coast

0.83

 Cape

0.83

Nam

0.82

reon

0.76

Rim

0.76

Activations Density 0.026%

mentions of geographical locations with the term "Cape" in them

occurrences of the word "Cape" in relation to various places

No Comments

No Known Activations