INDEX

Explanations

the word "guess" with various levels of emphasis or context

oai_token-act-pair · gpt-3.5-turbo

phrases that involve the act of guessing

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

interrupted

-0.76

ksh

-0.76

enture

-0.71

iences

-0.71

ĸļ

-0.70

umbai

-0.70

oran

-0.66

helle

-0.63

andals

-0.63

contained

-0.62

POSITIVE LOGITS

 guesses

1.01

 guess

0.91

 guessed

0.88

 guessing

0.84

correct

0.81

 Guess

0.79

 correctly

0.78

 whence

0.77

work

0.77

 incorrectly

0.71

Activations Density 0.039%

the word "guess" with various levels of emphasis or context

phrases that involve the act of guessing

No Comments

No Known Activations