INDEX

Explanations

sentences containing the word "here"

oai_token-act-pair · gpt-3.5-turbo

references to the term "here."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

asting

-0.61

jug

-0.57

 absorbing

-0.54

 sway

-0.54

oppable

-0.54

 meticulous

-0.54

 scoop

-0.53

ripp

-0.53

 scare

-0.53

san

-0.53

POSITIVE LOGITS

abouts

1.19

upon

1.04

fore

0.97

here

0.95

tics

0.94

soever

0.83

theless

0.81

tical

0.79

iton

0.78

after

0.77

Activations Density 0.011%

sentences containing the word "here"

references to the term "here."

No Comments

No Known Activations