INDEX

Explanations

phrases related to the act of excusing oneself or being excused from something

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "use" or "utilization."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 epoch

-0.82

izoph

-0.79

pton

-0.71

arat

-0.70

lake

-0.67

ohm

-0.64

 batt

-0.64

arger

-0.60

 neigh

-0.60

merce

-0.58

POSITIVE LOGITS

uses

0.89

iously

0.85

used

0.80

ãĤ¼

0.77

ibilities

0.75

iveness

0.75

umn

0.73

 sidx

0.71

ibles

0.70

umes

0.69

Activations Density 0.018%

phrases related to the act of excusing oneself or being excused from something

words related to the concept of "use" or "utilization."

No Comments

No Known Activations