INDEX

Explanations

search-related terms, particularly focusing on the word "search" and its occurrences in different contexts

oai_token-act-pair · gpt-3.5-turbo

references to search engines and search-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ework

-0.82

pered

-0.77

 Leilan

-0.76

 ceremonial

-0.65

iations

-0.64

HAEL

-0.64

ioch

-0.63

owe

-0.63

enei

-0.62

 sacr

-0.61

POSITIVE LOGITS

 queries

1.09

 query

1.04

results

0.94

 algorithm

0.92

query

0.92

 results

0.90

 keywords

0.89

ãĤ¦ãĤ¹

0.88

 algorithms

0.87

Results

0.86

Activations Density 0.059%

search-related terms, particularly focusing on the word "search" and its occurrences in different contexts

references to search engines and search-related activities

No Comments

No Known Activations