INDEX

Explanations

descriptions of objects being equipped with specific features or items

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "equipped" and its related contexts

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

-0.72

birth

-0.71

amin

-0.66

ctrl

-0.66

 Origin

-0.65

cott

-0.65

PU

-0.64

aji

-0.64

ort

-0.63

noon

-0.61

POSITIVE LOGITS

oise

0.87

bod

0.77

aton

0.76

 equipped

0.73

 showc

0.73

illance

0.73

equipped

0.72

odied

0.71

 racks

0.71

folios

0.71

Activations Density 0.021%

descriptions of objects being equipped with specific features or items

occurrences of the word "equipped" and its related contexts

No Comments

No Known Activations