INDEX

Explanations

statistics and percentages

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "all" in various contexts

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

SHIP

-0.67

lav

-0.65

icter

-0.63

dt

-0.63

zik

-0.62

 tremend

-0.62

 Maduro

-0.61

 simultane

-0.60

 separat

-0.60

 impunity

-0.60

POSITIVE LOGITS

ocating

1.22

ocated

0.93

ocation

0.93

iter

0.92

igators

0.92

owing

0.88

ogene

0.85

igator

0.84

uv

0.84

ocations

0.83

Activations Density 0.025%

statistics and percentages

occurrences of the word "all" in various contexts

No Comments

No Known Activations