INDEX

Explanations

instances of the word "Us" with varying importance, potentially based on context

oai_token-act-pair · gpt-3.5-turbo

the term "Us" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

served

-0.82

*/(

-0.68

 BART

-0.62

paralle

-0.62

 chaired

-0.61

committee

-0.61

 secrecy

-0.59

seed

-0.59

 jail

-0.59

POSITIVE LOGITS

agi

1.00

Us

0.96

AGES

0.90

ween

0.86

awar

0.86

urers

0.84

ages

0.84

ubi

0.82

ability

0.82

selves

0.80

Activations Density 0.004%

instances of the word "Us" with varying importance, potentially based on context

the term "Us" in various contexts

No Comments

No Known Activations