INDEX

Explanations

the word "see" when paired with a verb expressing observation, prediction, or opinion

oai_token-act-pair · gpt-3.5-turbo

instances of the word "see" indicating subjective perceptions or viewpoints

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

inion

-0.69

trap

-0.69

 agric

-0.68

wake

-0.68

ascal

-0.67

emetery

-0.65

afort

-0.64

 anthem

-0.63

adding

-0.62

 dispers

-0.62

POSITIVE LOGITS

dust

0.87

Ĺ

0.79

eye

0.79

 parallels

0.79

lights

0.79

awed

0.78

peak

0.78

 visions

0.77

fit

0.73

ById

0.73

Activations Density 0.080%

the word "see" when paired with a verb expressing observation, prediction, or opinion

instances of the word "see" indicating subjective perceptions or viewpoints

No Comments

No Known Activations