INDEX

Explanations

phrases related to discovery or uncovering

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "found" in various contexts, particularly related to discoveries or revelations

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

idium

-0.85

 horm

-0.78

inion

-0.71

adan

-0.63

puff

-0.61

VP

-0.61

 tert

-0.61

raviolet

-0.60

 paced

-0.60

 privile

-0.60

POSITIVE LOGITS

gdala

0.77

éļ

0.75

 è£ıè

0.72

Ô

0.72

andals

0.72

ymes

0.71

EMS

0.71

 caches

0.69

 clues

0.68

ãĤ¤ãĥĪ

0.68

Activations Density 0.049%

phrases related to discovery or uncovering

occurrences of the word "found" in various contexts, particularly related to discoveries or revelations

No Comments

No Known Activations