INDEX

Explanations

terms related to technological tasks and processes

oai_token-act-pair · gpt-3.5-turbo

gerunds and present participles

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Mub

-0.67

 pronounced

-0.62

bra

-0.61

ega

-0.61

 foresee

-0.61

si

-0.60

omet

-0.59

ombat

-0.59

ï¿½ï¿½ï¿½ï¿½

-0.58

 ï¿½ï¿½ï¿½ï¿½

-0.57

POSITIVE LOGITS

redients

1.23

tons

1.00

bats

0.90

HAM

0.85

GGGGGGGG

0.80

uez

0.78

athering

0.78

ton

0.75

allery

0.72

apore

0.66

Activations Density 0.143%

terms related to technological tasks and processes

gerunds and present participles

No Comments

No Known Activations