INDEX

Explanations

words related to financial transactions or documentation, particularly those related to checks

oai_token-act-pair · gpt-3.5-turbo

variations of the word "que" in different contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 mindfulness

-0.60

 Improvement

-0.59

 forecasts

-0.59

 trusts

-0.59

RSS

-0.59

 setbacks

-0.59

 stages

-0.58

 gauge

-0.58

 Lithuania

-0.58

 Components

-0.57

POSITIVE LOGITS

que

4.91

ques

3.11

Que

1.75

qu

1.50

quet

1.36

1.33

qua

1.28

quer

1.25

aque

1.22

quin

1.21

Activations Density 0.012%

words related to financial transactions or documentation, particularly those related to checks

variations of the word "que" in different contexts

No Comments

No Known Activations