INDEX

Explanations

the word "care" and words related to the concept of care

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "care" in various contexts

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

ãĥĥãĥī

-0.82

kered

-0.77

ãĥ³ãĤ¸

-0.73

 rand

-0.70

 bluff

-0.65

akedown

-0.61

è¦ļéĨĴ

-0.61

awe

-0.59

Ze

-0.58

 Monstrous

-0.58

POSITIVE LOGITS

taker

1.59

giving

1.34

taking

1.13

fully

1.07

lessly

1.05

lessness

1.04

tta

1.04

ful

0.97

ening

0.95

fulness

0.93

Activations Density 0.031%

the word "care" and words related to the concept of care

occurrences of the word "care" in various contexts

No Comments

No Known Activations