INDEX

Explanations

words related to physical chains or sequences of events

oai_token-act-pair · gpt-3.5-turbo

references to "chain" as a recurring concept 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

sburg

-0.80

 "$:/

-0.77

racuse

-0.77

iator

-0.74

iage

-0.73

issance

-0.73

farious

-0.71

mble

-0.70

gdala

-0.70

orescent

-0.70

POSITIVE LOGITS

alysis

0.96

chain

0.91

mail

0.90

loader

0.87

wal

0.86

 chain

0.84

breaker

0.82

 Reaction

0.81

breaking

0.80

params

0.80

Activations Density 0.040%

words related to physical chains or sequences of events

references to "chain" as a recurring concept in various contexts

No Comments

No Known Activations