INDEX

Explanations

terms related to the word "Strat"

oai_token-act-pair · gpt-3.5-turbo

references to strategic concepts and entities

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

VID

-0.77

Interstitial

-0.68

ternity

-0.65

hyde

-0.65

fulness

-0.65

 Recall

-0.65

ãĥ¼ãĥ³

-0.64

give

-0.64

 Clicker

-0.63

çīĪ

-0.61

POSITIVE LOGITS

osp

1.09

osphere

0.97

raph

0.93

itud

0.92

igon

0.91

igraph

0.91

itudes

0.87

igm

0.84

orman

0.80

ently

0.79

Activations Density 0.013%

terms related to the word "Strat"

references to strategic concepts and entities

No Comments

No Known Activations