INDEX

Explanations

phrases starting with "We" and referring to collective experiences or actions

oai_token-act-pair · gpt-3.5-turbo

repeated use of the word "we" indicating a collective or group perspective

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

forms

-0.76

¿½

-0.62

 persists

-0.61

states

-0.60

 arises

-0.57

amide

-0.56

 entails

-0.56

ocl

-0.56

cum

-0.55

 constitutes

-0.54

POSITIVE LOGITS

're

1.36

ird

1.21

asel

1.10

've

1.08

athered

1.08

 weren

1.07

bsite

1.03

eks

1.02

ighed

1.01

'll

1.01

Activations Density 0.208%

phrases starting with "We" and referring to collective experiences or actions

repeated use of the word "we" indicating a collective or group perspective

No Comments

No Known Activations