INDEX

Explanations

abbreviations or specific terms related to medical conditions such as "LSU" and "LS"

oai_token-act-pair · gpt-3.5-turbo

occurrences of 'LS' and related terms that indicate logistical or statistical information

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

ãĥĥãĥĪ

-0.76

 stoked

-0.72

natal

-0.71

shi

-0.67

fare

-0.65

stroke

-0.62

oslav

-0.62

 living

-0.60

ously

-0.60

framework

-0.60

POSITIVE LOGITS

heed

1.11

aturday

1.04

earch

0.90

gow

0.89

LS

0.89

ensitive

0.88

eeds

0.88

aunders

0.86

uable

0.86

ued

0.85

Activations Density 0.023%

abbreviations or specific terms related to medical conditions such as "LSU" and "LS"

occurrences of 'LS' and related terms that indicate logistical or statistical information

No Comments

No Known Activations