INDEX

Explanations

phrases indicating a contrast or contradiction

oai_token-act-pair · gpt-3.5-turbo

phrases that highlight contradictory actions or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

quartered

-0.62

 skeletal

-0.58

ampa

-0.57

ooming

-0.57

uman

-0.56

ropolis

-0.55

 Argon

-0.55

aturated

-0.54

celer

-0.54

nces

-0.54

POSITIVE LOGITS

 Advice

0.70

 preferably

0.70

 PLEASE

0.65

 Pledge

0.63

 Leave

0.63

 politely

0.62

Write

0.61

phrase

0.60

gently

0.60

ãĤ®

0.59

Activations Density 1.043%

phrases indicating a contrast or contradiction

phrases that highlight contradictory actions or behaviors

No Comments

No Known Activations