INDEX

Explanations

phrases related to consequences or impacts

oai_token-act-pair · gpt-3.5-turbo

the word "that" 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

Desk

-0.70

gur

-0.68

lander

-0.66

ron

-0.65

vre

-0.65

roth

-0.65

river

-0.64

ji

-0.62

ilet

-0.61

uty

-0.61

POSITIVE LOGITS

 accompanies

0.84

 consumes

0.83

 surrounds

0.83

 spawned

0.82

 caused

0.80

 mattered

0.79

 arose

0.78

 preceded

0.77

 cumbers

0.74

 THEY

0.73

Activations Density 0.227%

phrases related to consequences or impacts

the word "that" in various contexts

No Comments

No Known Activations