INDEX

Explanations

words related to losing or being lost

oai_token-act-pair · gpt-3.5-turbo

instances of the word "lost" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

osi

-0.77

ansky

-0.75

ibli

-0.70

OPLE

-0.70

enegger

-0.70

oples

-0.68

cius

-0.68

anwhile

-0.65

ffield

-0.64

NVIDIA

-0.62

POSITIVE LOGITS

itive

0.81

luster

0.75

 lost

0.71

touch

0.71

 Password

0.70

hearted

0.70

 overboard

0.69

 aversion

0.69

esses

0.69

 Souls

0.68

Activations Density 0.030%

words related to losing or being lost

instances of the word "lost" in various contexts

No Comments

No Known Activations