INDEX

Explanations

the word "pretend" in various contexts

oai_token-act-pair · gpt-3.5-turbo

expressions related to the concept of pretending or faking

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

âĨĳ

-0.72

OTOS

-0.66

burn

-0.65

Bio

-0.61

Parameter

-0.61

srf

-0.61

hner

-0.60

lean

-0.59

Issue

-0.59

Region

-0.59

POSITIVE LOGITS

 forgot

0.77

 innocence

0.76

entious

0.75

ulence

0.74

 pretend

0.72

pas

0.71

arial

0.69

cha

0.69

 pretending

0.67

bab

0.67

Activations Density 0.011%

the word "pretend" in various contexts

expressions related to the concept of pretending or faking

No Comments

No Known Activations