INDEX

Explanations

the word "mere" in contexts where it represents something small or insignificant

oai_token-act-pair · gpt-3.5-turbo

instances of the word "mere" along with phrases indicating insignificance or triviality

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 admired

-0.79

lain

-0.70

 disliked

-0.69

 disbanded

-0.69

anwhile

-0.66

airs

-0.65

 Shining

-0.65

 shaped

-0.65

 feared

-0.64

 Indies

-0.64

POSITIVE LOGITS

 mortals

1.05

 mere

1.04

 guiActiveUn

0.97

pas

0.83

cms

0.82

 mortal

0.79

ãĥ¬

0.78

\\\\\\\\

0.77

stros

0.76

reet

0.73

Activations Density 0.006%

the word "mere" in contexts where it represents something small or insignificant

instances of the word "mere" along with phrases indicating insignificance or triviality

No Comments

No Known Activations