INDEX

Explanations

phrases related to specific locations or organizations

oai_token-act-pair · gpt-3.5-turbo

the presence of the word "des" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iary

-0.72

 tipped

-0.69

 Holmes

-0.62

Disclaimer

-0.61

reads

-0.61

 Keynes

-0.60

Behind

-0.60

runner

-0.60

estone

-0.58

 favorites

-0.57

POSITIVE LOGITS

 congr

0.90

plet

0.89

ription

0.81

ignt

0.79

ugar

0.79

 masse

0.77

pite

0.75

ider

0.74

lict

0.72

icc

0.72

Activations Density 0.007%

phrases related to specific locations or organizations

the presence of the word "des" in various contexts

No Comments

No Known Activations