INDEX

Explanations

adjectives describing characteristics or qualities of objects or actions

oai_token-act-pair · gpt-3.5-turbo

adjectives describing various qualities or states of objects and experiences

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

moon

-0.63

¬¼

-0.63

thood

-0.62

xit

-0.62

©¶æ

-0.61

usalem

-0.61

CHA

-0.59

aleb

-0.58

rera

-0.57

bey

-0.57

POSITIVE LOGITS

 compared

0.91

 insofar

0.84

 nowadays

0.82

 sounding

0.81

 indeed

0.80

––

0.77

(~

0.75

 anyway

0.75

 anyways

0.74

looking

0.72

Activations Density 0.284%

adjectives describing characteristics or qualities of objects or actions

adjectives describing various qualities or states of objects and experiences

No Comments

No Known Activations