INDEX

Explanations

phrases related to physical or emotional trauma, particularly involving force or being forced

oai_token-act-pair · gpt-3.5-turbo

references to personal autonomy and agency in the context of gender

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

conservancy

-0.69

ugen

-0.67

 satell

-0.65

 cooled

-0.63

 tremend

-0.63

Ã±a

-0.62

iasm

-0.61

stood

-0.60

 hindsight

-0.60

 feasibility

-0.60

POSITIVE LOGITS

tackle

0.78

Quotes

0.69

 Bender

0.69

days

0.68

Topics

0.68

catch

0.66

 Steal

0.65

ebin

0.65

 respectively

0.64

rities

0.64

Activations Density 0.000%

phrases related to physical or emotional trauma, particularly involving force or being forced

references to personal autonomy and agency in the context of gender

No Comments

No Known Activations