INDEX

Explanations

words related to being a byproduct or end result of a process or activity

oai_token-act-pair · gpt-3.5-turbo

terms related to economic measurements and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

conservancy

-0.69

 needle

-0.69

bat

-0.66

 coil

-0.64

 Tarant

-0.61

 mosqu

-0.61

drive

-0.61

 stake

-0.61

 Telescope

-0.60

Winged

-0.60

POSITIVE LOGITS

ivity

1.30

iveness

1.19

ively

1.18

ivities

1.13

ions

0.99

iation

0.91

ivist

0.88

ivation

0.88

ivism

0.86

ivated

0.82

Activations Density 0.023%

words related to being a byproduct or end result of a process or activity

terms related to economic measurements and statistics

No Comments

No Known Activations