INDEX

Explanations

adjectives that describe actions or characteristics with a strong emotional connotation

oai_token-act-pair · gpt-3.5-turbo

strong adjectives and descriptors indicating emotional intensity or criticism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

slave

-0.77

orks

-0.76

agara

-0.75

hov

-0.74

osponsors

-0.74

 Alive

-0.73

laus

-0.72

upiter

-0.71

ulhu

-0.71

conservancy

-0.70

POSITIVE LOGITS

 rebuke

1.33

 remarks

1.32

 rebutt

1.26

 critique

1.19

 statement

1.18

 critiques

1.16

 speeches

1.15

 criticism

1.15

 reply

1.14

 statements

1.14

Activations Density 0.147%

adjectives that describe actions or characteristics with a strong emotional connotation

strong adjectives and descriptors indicating emotional intensity or criticism

No Comments

No Known Activations